Radiological studies of fetal alcohol spectrum disorders in humans and animal models: An updated comprehensive review

Magnetic Resonance Imaging 43 (2017) 10–26

Contents lists available at ScienceDirect

Magnetic Resonance Imaging journal homepage: www.mrijournal.com

Radiological studies of fetal alcohol spectrum disorders in humans and animal models: An updated comprehensive review Van T. Nguyen a,b,1, Suyinn Chong c,d, Quang M. Tieng a, Karine Mardon a, Graham J. Galloway a,d, Nyoman D. Kurniawan a,⁎ a

Centre for Advanced Imaging, The University of Queensland, Brisbane, Queensland, Australia Hanoi University of Science and Technology, Hanoi, Vietnam Mater Research Institute, The University of Queensland, Brisbane, Queensland, Australia d Translational Research Institute, Brisbane, Queensland, Australia b c

a r t i c l e

i n f o

Article history: Received 28 August 2016 Received in revised form 19 June 2017 Accepted 19 June 2017

Keywords: FASD Radiology Humans Animals Prenatal alcohol exposure

a b s t r a c t Fetal Alcohol Spectrum Disorders encompass a wide range of birth defects in children born to mothers who consumed alcohol during pregnancy. Typical mental impairments in FASD include difficulties in life adaptation and learning and memory, deficits in attention, visuospatial skills, language and speech disabilities, mood disorders and motor disabilities. Multimodal imaging methods have enabled in vivo studies of the teratogenic effects of alcohol on the central nervous system, giving more insight into the FASD phenotype. This paper offers an up-todate comprehensive review of radiological findings in the central nervous system in studies of prenatal alcohol exposure in both humans and translational animal models, including Magnetic Resonance Imaging, Computed Tomography, Positron Emission Tomography, Single Photon Emission Tomography and Ultrasonography. © 2017 Elsevier Inc. All rights reserved.

Contents 1. 2.

Introduction to FASD . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.1. Neurodevelopmental and neurobehavioral abnormalities in FASD . . . . . . . . . . . . . . . . . . . . Radiological studies of humans with FASD . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1. Detection of brain structural changes using structural MRI . . . . . . . . . . . . . . . . . . . . . . . 2.1.1. Alterations in gray matter . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1.2. Alterations in white matter . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1.3. Links between abnormal brain structures and cognition in FASD . . . . . . . . . . . . . . . . 2.2. Brain microstructural changes revealed by DTI . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.3. Correlations between DTI and functional changes . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.4. Detection of brain metabolite changes using MRS . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.5. Functional brain imaging . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.5.1. Altered brain network recruitment and activation in task performance revealed by task-based fMRI Working memory . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Visual attention . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Verbal memory function . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Motor control and response inhibition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . .

11 11 12 12 12 12 13 14 14 14 15 15 15 15 16 16

Abbreviations: AD, Axial diffusivity; ADHD, Attention deficit hyperactivity disorder; ARBD, Alcohol-related birth defects; ARND, Alcohol-related neurodevelopmental disorders; BOLD, Blood-oxygen level dependent; CNS, Central nervous system; CT, Computed tomography; DMN, Default mode network; DTI, Diffusion tensor imaging; FA, Fractional anisotropy; FAS, Fetal alcohol syndrome; FASD, Fetal alcohol spectrum disorders; GD, Gestational day; ILF, Inferior longitudinal fasciculus; MD, Mean diffusivity; MRI, Magnetic resonance imaging; PAE, Prenatal alcohol exposure; PET, Positron emission tomography; PFAS, Partial FAS; RD, Radial diffusivity; SES, Socioeconomic status; SLF, Superior longitudinal fasciculus; SPECT, Single photon emission tomography. ⁎ Corresponding author at: Building 57 Research Rd, St Lucia, QLD 4072, Australia. E-mail addresses: [email protected] (V.T. Nguyen), [email protected] (N.D. Kurniawan). 1 Primary author.

http://dx.doi.org/10.1016/j.mri.2017.06.012 0730-725X/© 2017 Elsevier Inc. All rights reserved.

V.T. Nguyen et al. / Magnetic Resonance Imaging 43 (2017) 10–26

Number processing function . . . . . . . . . . . . . . . . . . 2.5.2. Brain network changes revealed by resting-state fMRI . . 2.5.3. PET and SPECT . . . . . . . . . . . . . . . . . . . . 2.6. Ultrasonography . . . . . . . . . . . . . . . . . . . . . . . 2.7. Insights from radiological studies of humans with FASD . . . . . . 3. Radiological studies of translational animal models of PAE . . . . . . . . 3.1. Non-human primate PAE studies . . . . . . . . . . . . . . . . 3.1.1. An early non-human primate MRS study . . . . . . . . 3.1.2. Non-human primate PET studies . . . . . . . . . . . . 3.2. Other animal models of PAE . . . . . . . . . . . . . . . . . . 3.2.1. Mouse models of acute high dose of PAE . . . . . . . . 3.2.2. Animal models of chronic high dose PAE . . . . . . . . 3.2.3. Rodent models of chronic moderate dose of PAE . . . . . 3.2.4. Rodent models of chronic low dose of PAE . . . . . . . . 3.2.5. Insights from radiological studies of animal models of PAE 3.3. Comparison of PAE effects in humans and animal models . . . . . 4. Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Funding . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Acknowledgement . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

1. Introduction to FASD The effect of maternal alcohol consumption on fetal development was first described in 1968 and the term Fetal Alcohol Syndrome (FAS) was first used in 1973 to describe the adverse effects observed in children born to alcoholic mothers [1]. FAS was defined by four criteria: (1) prenatal alcohol exposure (PAE), (2) growth retardation, (3) facial abnormalities and (4) central nervous system (CNS) deficits with or without structural brain anomalies [2]. Distinct craniofacial characteristics of FAS include smaller head size, lower nasal bridge, minor ear abnormalities, flattened philtrum, epicanthal folds, short palpebral fissure, flat midface, short nose, and thinner upper lip [3] (Fig. 1). FAS is the most severe manifestation of a wider spectrum of disorders, termed Fetal Alcohol Spectrum Disorders (FASD). FASD includes less severe subtypes called partial FAS (PFAS), Alcohol Related Birth Defects (ARBD) and Alcohol Related Neurodevelopmental Disorders (ARND) [4]. PFAS refers to individuals who exhibit at least two of the three key facial anomalies: flattened philtrum, short palpebral fissure and thin upper lip; and one of the following: small head circumference, growth retardation, or neurobehavioral impairments. ARBD refers to individuals with a history of PAE and congenital physical anomalies. ARND, the most common subtype in the spectrum [5], refers to individuals with PAE-induced CNS abnormalities without facial anomalies [6–8]. FAS/PFAS is called dysmorphic FASD condition while ARND nondysmorphic FASD condition [9]. There is a wide range in the reported

. . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . .

11

. . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . .

16 16 16 16 17 17 17 17 20 20 20 22 22 22 23 23 23 23 23 24

prevalence of these disorders, including 55.42 (per 1000) for FAS, 20.25 for ARND and 113.22 for FASD in South Africa, and 1.33, 9.07 and up to 47.13 respectively in the United States [10]. Children with heavy PAE typically have lower birth-weight than normal and demonstrate low weight during development into adolescence and adulthood [11,12], while those intermediately exposed may show a growth catchup during development [12]. The weight of individuals with FASD is also disproportionately low compared to their height [11]. Cardiac, skeletal, renal, ocular, or auditory anomalies have also been described in individuals with ARBD [7]. Clinical examinations and radiography have identified skeletal defects in a group of children diagnosed with FAS, including bilateral and unilateral fusion of the capitae and hamate bones in the carpus and bilateral radio-ulnar fusion at the proximal end (Fig. 2) [13]. 1.1. Neurodevelopmental and neurobehavioral abnormalities in FASD CNS dysfunction in individuals with FASD is manifested through mild to severe cognitive deficiencies and neurobehavioral disorders [8]. Typical neurobehavioral and neuropsychological deficiencies include attention deficit and hyperactivity disorder (ADHD), difficulties in life adaptation, difficulties in learning and memory, deficits in visuospatial skills, language and speech disabilities, mood disorders and motor disabilities [14].

Fig. 1. Typical facial phenotype of FAS. Adapted from [180].

12

V.T. Nguyen et al. / Magnetic Resonance Imaging 43 (2017) 10–26

Fig. 2. Skeletal abnormalities in FASD. Synostosis of capitae and hamate bones in the carpus (asterisk) (a) and radio-ulnar fusion of the arm (arrow) (b). Adapted from [13].

Early studies relied on postmortem examinations which revealed dysmorphic brain macrostructures, such as cortical and cerebellar heterotopias, smaller and hypoplastic cerebellum, absence of pons and poorly formed medulla [15]. However, with the advent of modern radiological imaging techniques, non-invasive studies of brain abnormalities are now possible and serial imaging can be used to monitor development. The most common modalities used to study FASD are Magnetic Resonance (MR)-based imaging, including structural MR Imaging (sMRI), Magnetic Resonance Spectroscopy (MRS), functional MRI (fMRI) and Diffusion Tensor Imaging (DTI). Less commonly used are Computed Tomography (CT), Ultrasonography, Positron Emission Tomography (PET) and Single Photon Emission Tomography (SPECT). Although functional imaging like fMRI, PET and SPECT are advantageous in that they can reveal functional changes when gross structural changes are absent or not detectable, only few studies have utilized PET and SPECT due to their high cost, difficulties in experimental setup and dependence on the availability of the radiotracers [16]. This paper aims to comprehensively review the methodologies and findings of radiological studies in both humans with FASD and translational animal models of PAE. 2. Radiological studies of humans with FASD A large number of imaging studies have been performed in humans with FASD over the past few decades reporting a wide range of PAEinduced outcomes, which will be discussed in the following sections. 2.1. Detection of brain structural changes using structural MRI Structural MRI (sMRI) is the most widely used MR-based protocol, with its ability to pinpoint gross brain volume changes and shape deformity. Common observations from FASD studies using sMRI include smaller brain size [5,17–23], in line with volume reductions of total gray matter [5,18,24,25], white matter [18,26] and changes in more distinct brain regions. 2.1.1. Alterations in gray matter In children and adolescents with FASD, gray matter alterations have been detected in various cortical lobes including the frontal [5,20], temporal [5,17,27], parietal and occipital lobes [17,27]. The degrees of volume reduction can vary for different lobes, with some studies observing that the parietal lobe showed greater volume reduction than the temporal and occipital lobes [17,27]. Reduced cortical volumes have also been observed in the left cuneus cortex, the lingual and right inferior temporal gyri [28]; the left parietooccipital and right anterior prefrontal cortices [29]; the left cingulate gyrus, the right middle frontal and temporal gyri [30]; the left superior frontal gyrus [25]; and the middle frontal, supramarginal and inferior parietal regions [31].

Children and adolescents with FASD exhibited reduced cortical thickness globally [18,21], as well as bilaterally in the middle frontal lobe, the large central area, the lateral inferior temporal and occipital lobes [32], the left orbitofrontal region [33], the occipitotemporal and parietal regions [34], the bilateral superior parietal, right postcentral, left precentral and lateral occipital regions [35]. Additionally, reduced cortical surfaces were observed in the anterior cingulate cortex [36], the right temporal lobe, the right superior temporal gyrus and the region between the right temporal and occipital cortices [5], and mid-sagittal anterior and posterior inferior lobes [37]. Reduced cortical folding, both with and without microcephaly, was observed in children with heavy PAE [38]. In contrast to the commonly observed reductions, some studies have reported increases in cortical thickness in children and young adults with FASD in the lateral occipital regions [39], frontal, temporal and parietal regions [40]. Gray matter density was found increased in the parietal region [41,42], unilaterally in the peri-Sylvian cortices of the temporal and parietal lobes [43], and the frontal and occipital temporal regions [42]. Cortical volumes were increased in the anterior and posterior cingulate regions and in the peri-Sylvian cortices [29]. The rate and trajectory of cortical development were also different between individuals with FASD and healthy peers. In children with PAE, the cortices were thinner at one time-point in childhood, and displayed a slower rate of thinning over an examined developmental period in multiple regions, particularly the medial frontal and parietal regions, compared to healthy children [21]. In another study, while healthy children exhibited prolonged cortical volume increase followed by cortical volume decrease, children with PAE showed cortical volume decrease only [26]. The right and left transverse temporal regions in adolescents with FASD exhibited significant linear volume increases over a two year timecourse while typically developing peers did not [44]. Adolescents with heavy PAE also showed a lack of normal asymmetry in gray matter density in the posterior temporal lobes compared to healthy individuals [27]. In line with changes in cortical regions and surface area, significant volume changes were observed in distinct brain structures as well. In children and adolescents with FASD, large volume reductions were seen in the cerebrum [45], cerebellum (Fig. 3) [37,45–48], cerebellar vermis [20,49], caudate, putamen, hippocampus [18,19,28–30,47,50, 51], thalamus [18,29,51,52], basal ganglia [45,48,52–54], globus pallidus and amygdala [18]. Volume reduction and structural dysmorphology co-occurred in the case of the cerebellar vermis [49], but structural dysmorphology and asymmetry were present in the absence of volume changes in the case of the hippocampus and caudate nucleus [55,56]. Several studies reported wider cortical sulci [19,47,50]. 2.1.2. Alterations in white matter Similar to cortical gray matter changes, reductions in white matter volume were observed globally [18,26], as well as around the occipital

V.T. Nguyen et al. / Magnetic Resonance Imaging 43 (2017) 10–26

13

Fig. 3. Dysmorphia of the cerebellar vermian sulci (arrows): normal (a); slight widening (b); moderate widening (c) and heavy widening (d). Adapted from [47].

horns [48], in the left parietal and temporal regions [41,43], the ventral occipital temporal region [42], and the posterior cingulate region [57]. Midline brain structures exhibited high vulnerability to the adverse effects of PAE. Abnormalities included hippocampal commissure agenesis [23], callosal malformation, cavum septi pellucidi and cavum vergae [58]. The corpus callosum, the biggest white matter structure in the brain, which connects the two cerebral hemispheres, has been examined in many PAE investigations. Following PAE, the corpus callosum was found to be displaced in the inferior–superior and anterior–posterior directions [59]; decreased in size [23,31,45,47,48,52,58,60–63] or completely absent [45] (Fig. 4), but also thickened [61]. Discrete regions of the corpus callosum appeared to be affected differentially. Volume reductions were seen in the isthmus and splenium [59,64], but not in other parts of the corpus callosum. The superior posterior region between the

isthmus and splenium was also more severely displaced as compared to the more anterior part [59]. It should be noted that the decreased thickness and area of the corpus callosum were associated with decreased palpebral fissure length, a FASD facial characteristic [58,63], and the severity of midline structural malformations correlated with the severity of facial anomalies in individuals with FASD [58]. 2.1.3. Links between abnormal brain structures and cognition in FASD In individuals with FASD, changes in the volume and thickness of various brain regions have been linked to altered cognitive performance. For example, a smaller caudate nucleus, a structure that is essential in learning and memory processing and storing [65], was found to strongly correlate with poor cognitive performance and verbal learning ability [66]. Poor verbal learning ability was also correlated with the amount of

Fig. 4. Hypogenesis (a) and agenesis (b) of the corpus callosum in children with FAS. Adapted from [20].

14

V.T. Nguyen et al. / Magnetic Resonance Imaging 43 (2017) 10–26

displacement of the corpus callosum [59] and the cerebellar vermis [49]. In individuals with heavy PAE, a thinner dorsal frontal cortex was associated with poorer performance in verbal learning; however, the correlation between thinner frontal, parietal and temporal cortices and better visuospatial functioning did not exist as in healthy peers [39]. Cortical thickness and cognitive scores showed a positive correlation in the case of high dose alcohol exposure, but showed a negative correlation in the case of low dose exposure [33]. A smaller caudal anterior cingulate cortex in the right hemisphere, a region involved in inhibition processing and error monitoring, was related to longer time spent on inhibiting a response [36]. Volume reductions of the isthmus, splenium or the whole corpus callosum were associated with increased interhemispheric transfer-related errors in finger localization tests and executive functioning [61,64,67], while volume reduction of the hippocampus mediated memory performance of PAE group in a verbal learning task [19]. 2.2. Brain microstructural changes revealed by DTI DTI involves the analysis of microscopic water molecule displacement in tissues using Gaussian distribution [68]. Typical measures in DTI quantification are fractional anisotropy (FA) and axial, radial and mean diffusivities (AD, RD and MD respectively). FA is a scalar parameter indicating the degree of anisotropy of diffusion processes and reflecting fiber density and axonal integrity [68]. AD refers to the diffusion along the principle axis (also called longitudinal diffusivity or parallel diffusivity), and a decrease in AD has been shown to associate with axonal degeneration or injury [69]. RD is the average of the diffusivities along two minor axes orthogonal to the main diffusion tensor axis (also called perpendicular diffusivity), and an increase in RD has been shown to associate with the degree of demyelination in white matter [69]. MD is a scalar measure indicating the total diffusion displacement within a voxel [69]. PAE has been found to result in alterations in all of these DTI-derived parameters in the brain. Consistent with sMRI-revealed abnormal midline structures, a number of DTI studies have reported decreased FA in the corpus callosum of children and young adults with FASD, both as a whole structure [70] and in different subregions including the splenium, posterior mid-body and isthmus [71–74]. Children and young adults with FASD also exhibited lower FA in the right cingulum, bilateral inferior and superior longitudinal fasciculi (ILF and SLF respectively) and left thalamus [74–76]; the frontal, parietal and occipital lobes [73,77]; the medial prefrontal cortex and posterior cingulate cortex [78]; the bilateral superior cerebellar peduncles [79,80] and the left middle peduncle [80]. The reduction in FA overlapped with an increase in RD in the medial prefrontal and posterior cingulate cortices [78], the superior cerebellar peduncles [79], left ILF and isthmus [74], and the left middle peduncle [79,80]; whereas increased FA was only observed in the bilateral globus pallidus [75]. In children and young adults with FASD, higher MD was seen in the bilateral inferior frontooccipital fasciculi, SLF, left ILF, corticospinal tracts, globus pallidus, right putamen and right thalamus [74–76]; left superior frontal lobe and anterior limb of the internal capsule [73]; the splenium [74,81], isthmus [74,82] and whole corpus callosum [70,76]; the left middle and right inferior peduncles [79]; the cingulate, external capsule, uncinate fasciculus, superior corona radiate, arcuate fasciculus, frontal, fronto-central and central association fibers [76]; and the cingulum bundles that connect the medial prefrontal cortex and the posterior cingulate cortex [78]. Less commonly, lower MD was observed in the genu of corpus callosum [75], the right inferior frontal lobe and the right temporo-parieto-occipital junction [73]. MD decreased more steeply with age in the SLF and inferior frontooccipital fasciculi in FASD individuals [51], suggesting differences in the rates of brain maturity between children with FASD and healthy peers. It should be noted that increased RD accounted for changes in both FA and MD in the left middle peduncle [79,80], and the left ILF and isthmus [74,79] in PAE children. The combination of decreased FA and increased RD may reflect a decreased barrier to diffusion across the axon (such as reduced myelination, or demyelination) and/or reduced

axonal density, together with increased extracellular water content. The increase in extracellular water content can also be reflected through increased MD [83]. In neonates, AD appeared to be more sensitive to PAE than other diffusion parameters. Lower AD across the whole brain was associated with greater alcohol exposure [84], though the effect was strongest in the medial and inferior white matter. Significantly lower AD was also observed in the right SLF in PAE infants [85]. The changes in AD generally do not reflect demyelination but axonal state [69]. The decreased AD observed in the neonate brains may indicate axonal injury following PAE. Also noteworthy is that while many studies were carried out on individuals with confirmed syndromal FAS/PFAS, some studies have focused on heavily exposed nonsyndromal individuals (i.e. individuals with heavy PAE who do not show FAS/PFAS facial characteristics [74]) and showed that DTI could detect PAE-induced brain microstructural abnormalities in less severe cases [74,82]. Those studies, however, also showed that the level of exposure was associated with the level of alterations in diffusion parameters. Increased PAE was associated with decreases in FA and increases in RD [74,79,80] and increases in MD [74,79]. 2.3. Correlations between DTI and functional changes Alterations in diffusion parameters in individuals with FASD suggest abnormalities in white matter tract formation [69]. Changes in these parameters also showed a correlation with changes in specific brain functions such as learning [79,80,86], response inhibition [81,87]; and IQ [74]. Cerebellar peduncles are white matter tracts connecting the cerebellum and the brainstem. These structures control eye blink conditioning and saccadic eye movement, two paradigms that can be used as indicators for cognitive function [79,88]. The correlation between lower FA in the left cerebellum and increased saccadic reaction time [87]; and consistent correlation between lower FA in the cerebellar peduncles and poorer performance in eyeblink conditioning [79,80] suggest impaired neural transmission in these white matter tracts. In the infants with PAE, neurobehavioral scores exhibited a positive correlation with FA and a negative correlation with MD of the right inferior cerebellar peduncle [85]. In an antisaccade test, which is designed to assess response inhibition or cognitive control, normally developing children showed a negative correlation between FA of the corpus callosum and visual stimulus detection errors and a positive correlation between MD of the splenium and visual stimulus detection errors, while children with PAE did not exhibit such correlations [81]. These data suggest deficits in information transfer due to abnormal tract formation, resulting in failure in inhibition control (discussed in Section 2.5.1). 2.4. Detection of brain metabolite changes using MRS MRS is an in vivo technique used to examine concentrations of metabolites and neurotransmitters in the brain [89]. MRS can provide more sensitive indicators of underlying abnormalities in brain metabolism when structural aberrations are absent or undetectable [90]. Common brain metabolites of interest include N-acetyl aspartate (NAA), Creatine (Cr) and Choline (Cho). NAA is a neuron-specific marker and constitutes the most prominent peak in an MRS spectrum in healthy human brain tissue [91]. Changes in NAA are considered an indicator of neuronal dysfunction [89]. Cho is a marker of cellular membrane turnover. Cho levels in the brain are usually increased in the presence of tumors or inflammation [92]. Cr functions as the reservoir for high-energy phosphates in the cytosol of muscle and neurons, and appears to be generally unaffected by pathology and drug use, thus it is commonly used as a control for evaluating alterations in Cho and NAA levels [89]. In children and adolescents with PAE, considerable elevation in NAA/Cr ratio was observed in the caudate nucleus, which was attributable to a high level of NAA (rather than the decrease of Cr) in these regions [93]. In contrast, decreases in NAA were suggested to drive a lower NAA/Cr ratio in the frontal white matter, corpus callosum, thalamus and

V.T. Nguyen et al. / Magnetic Resonance Imaging 43 (2017) 10–26

cerebellar dentate nucleus, and lower NAA/Cho ratio in the parietal and frontal cortices [94]. Another study reported low NAA and Cho levels in the right frontal/parietal white matter region and the hippocampus, while Cr was unchanged, resulting in low NAA/Cr and Cho/Cr ratios in FASD group compared to control group [95]. A recent study detected lower levels of NAA and Cho in the deep cerebellar nuclei, a part of the cerebellum critical for cerebellar-mediated learning, in children born to mothers who were heavy drinkers around the time of conception and during pregnancy [96]. It was found that the decrease in NAA was more sensitive to exposure around conception, while Cho levels were more affected by exposure during pregnancy. In both cases, the metabolite levels were inversely proportional to maternal alcohol consumption, i.e. increased consumption was associated with a greater reduction in NAA and/or Cho. Due to the limited number of MRS studies in individuals with FASD, there is no clear consensus of the metabolite changes following PAE. Nonetheless, changes in brain metabolite markers may underpin structural alterations in FASD. For instance, the most consistent observation was reduced NAA levels, which may indicate significant neuronal loss [89] resulting in reduced brain volume (see Section 2.1), or impaired cellular signaling resulting in brain dysfunction (discussed in Section 2.5) [95]. It should be noted that NAA levels were increased in the caudate nucleus in one study [93]. Increased NAA levels are less common but have been observed in pathologies where a mutation in the enzyme aspartoacylase prevents the normal breakdown of NAA [91], and the resultant NAA buildup could, in turn, cause demyelination [97]. 2.5. Functional brain imaging Radiological functional imaging modalities typically used in FASD studies include fMRI, PET and SPECT. These modalities have been used to reveal impairments in brain functions in the absence of gross brain structural dysmorphology [16]. fMRI examines brain functional connectivity and activation during task performance (task-based fMRI) or at rest (resting-state fMRI) by measuring blood-oxygen-level dependent (BOLD) signals [98]. PET and SPECT can be used to examine metabolic activities of specific tissues or blood delivery to an organ to measure how well the organ is functioning [99]. Compared to PET and SPECT, fMRI is advantageous in that it does not require the use of a radioisotope, has higher spatial and temporal resolution and is more widely available [99]. PET and SPECT, on the other hand, are advantageous over fMRI in that they are not as prone to artifacts due to patient motion and local field inhomogeneity (air – tissue interfaces). PET and SPECT acquisition is acoustically less noisy, which is a major discomfort to participants in fMRI [99]. 2.5.1. Altered brain network recruitment and activation in task performance revealed by task-based fMRI Task-based fMRI studies have investigated a wide range of cognitive skills in individuals with PAE, including attention, learning, memory,

15

language, number processing and executive function, providing insight into stimulus-dependent changes in brain activation following PAE [98]. Attention deficits are proposed to underlie the increased reaction times and weaker task performance of individuals with FASD. Attention deficit hyperactivity disorder (ADHD) is present in many cases of FASD and was suggested as a core deficit in FASD [100,101]. However, the impairments in adaptive function or cognitive ability in individuals with ADHD and FASD appeared distinct in certain respects [102–104]. Working memory In the 1-back task, children with ARND showed broader activation of cortical regions compared to healthy children, which was correlated with decreased accuracy and increased response time variability, indicating increased effort to compensate for cognitive deficiency [103]. Despite similar task performance to children with ARND, children with ADHD showed weak activation of the same network, suggesting that children with ADHD relied on different brain regions compared to children with ARND [103]. In the N-back task, children with FASD performed significantly worse, with increased errors and reaction times, in comparison to their healthy peers [105]. fMRI detected considerably lower activation levels in the regions of greatest activity in a typically developing child, including the right inferior frontal gyrus, right posterior parietal lobe, right dorsolateral prefrontal cortex and right middle frontal gyrus [105], the superior frontal and parietal lobes [106], the temporal lobe [107], and the basal ganglia [108]. Subjects with FAS/PFAS also relied on a distinctly different subnetwork compared to normally developing peers and also to nonsyndromal heavily-exposed children [109]. Relative to the 0-back task, increased activations in the 1-back task were observed in the left inferior frontal gyrus in controls, and in the dorsal prefrontal cortex, caudate and putamen in nonsyndromal heavily-exposed children, while children with FAS/PFAS showed increased activity in two Crus I/lobule VI and lobule VIIB/VIIIA of the cerebellum and the inferior parietal cortex [109]. The activation increases from 0-back to 1-back task were significantly different between controls and PAE subjects in the inferior frontal region (higher in controls), and in the basal ganglia and dorsal prefrontal cortex (higher in PAE subjects) [109]. When participants had to identify and remember figure locations on a screen, adolescents with heavy PAE exhibited higher activation levels relative to baseline condition in the bilateral inferior, middle, and superior frontal gyri, superior temporal and occipital gyri, lentiform nuclei, and insulae as compared to controls [110,111]. Visual attention In a visual shape detection task to test sustained attention, young adults with FASD exhibited lower accuracy and longer reaction times compared to healthy peers. Higher activation of the superior occipital temporal area was observed, which may be attributable to the volume reduction in the ventral occipital temporal area in these subjects [42]. Attention deficits were also reflected through poor performance in

Fig. 5. Deactivation map of the default mode network. Functional connectivity between the medial prefrontal cortex (MPFC) and the posterior cingulate cortex (PCC) decreases with increasing severity of PAE: Control (a), non-dysmorphic PAE (b) and dysmorphic PAE (c). Bilateral cingulum bundles connecting MPFC and PCC (red bundles) (d). Adapted from [78].

16

V.T. Nguyen et al. / Magnetic Resonance Imaging 43 (2017) 10–26

search tasks that required participants to locate a character that first appeared alone then in an array [76]. In a simple search task where the target character reappeared together with three other characters (i.e. distractors whose color and shape were identical and different from the target's), adolescents with ARND had much more extensive activations in the dorsal frontoparietal network, which included the attention pathways, compared to the controls. Yet, when the search task became more complicated with the distractors either having the same color or shape as the target's, subjects with ARND showed only a small change in activation from the simpler task, while healthy subjects exhibited a substantial increase in cortical activation in the frontal, parietal, temporal and lateral occipital regions [76]. Verbal memory function In verbal paired associated learning to recall one word in a pair spoken earlier, children with FASD showed decreased functional activation in the left medial and posterior temporal regions; but increased activation in the right dorsal frontal cortex [112]. In a verbal working memory test where participants were required to recall words they had seen, children and adolescents with FASD showed increased activation in the left dorsal frontal and left inferior parietal cortices, and the bilateral posterior temporal regions compared to their healthy peers [113]. Motor control and response inhibition In children with PAE, the pattern of activation was altered in a task involving rhythmic and non-rhythmic finger tapping to assess motor control [114]. BOLD activation in the right crus I, vermis IV-VI and right lobule VI was higher in rhythmic than in non-rhythmic finger tapping task in healthy children; but this difference was decreased in the right crus I in both nonsydromal PAE children and children with FAS/PFAS. Additionally, children with FAS/PFAS showed the smallest increase in vermis IV-V. The difference in activation between rhythmic and non-rhythmic task was inversely correlated with maternal alcohol exposure (i.e. higher maternal alcohol intake resulted in smaller difference) [114]. When the task required the inhibition of a prepotent response rather than an action, children and adolescents with heavy PAE exhibited failures in inhibition control with altered BOLD response in the fronto–striatal circuit, a circuit involved in controlled inhibition of unwanted responses, compared to typically developing peers [115–117]. Number processing function Syndromal PAE children showed worse performance in a number processing task than healthy peers, with lower activation in the left superior and right inferior parietal regions and medial frontal gyrus, the regions related to arithmetic processing [118,119]. It should be noted that the performance of nonsyndromal PAE children was comparable to the controls, implying that poor performance increases with increased severity of the syndrome. The size of activated networks was also observed to be decreased with increasing severity of PAE [119]. In a number-related task requiring the judgment of number proximity, children with FASD either showed less activation in the intraparietal sulcus, one of the regions in the frontoparietal network involved in number processing [120], or used a different network to solve the problems [118]. In exact calculation tasks, brain activation in PAE children was more scattered and widespread compared to controls and included the cerebellar vermis and cortex, regions which have been found to be activated in adults engaged in challenging number processing problems [121]. 2.5.2. Brain network changes revealed by resting-state fMRI The study of basal brain activity at rest is called resting-state fMRI (rsfMRI), which investigates synchronous activations of remotely located brain regions in the absence of tasks [122]. The default mode network (DMN) is the most fundamental resting state network, comprising of the medial prefrontal cortex, the posterior lateral parietal cortex and the precuneus/posterior cingulate cortex [122]. DMN showed highly correlated activity at rest and its dysregulation was associated with poor

cognitive functioning in different psychiatric and neurological disorders [123]. Adults with FASD showed lower deactivation of the DMN during arithmetic tasks as well as lower resting functional connectivity between the medial prefrontal cortex and posterior cingulate cortex (Fig. 5) [78]. RsfMRI can measure global network efficiency, which reflects network capacity to communicate parallel information between remote brain regions. In children with FASD, global efficiency was decreased together with increased characteristic path length, which indicates less reliance on highly efficient hubs for long-distance communication but rather on multiple short jumps [124]. Regionally, lower inter-hemispheric functional connectivity was detected in children with FASD in the para-central region, the region connected by the posterior callosal fiber projection [125]. The children also showed no correlation between para-central functional connectivity and mean diffusivity, which were strongly correlated in control peers [125]. The absence of a correlation between the two parameters suggests disruption in the microstructures of the posterior region of the corpus callosum. Only one study to date has investigated the establishment of intrinsic resting brain network in neonates (2 to 4 weeks of age) [126]. Although a significant reduction in interhemispheric connectivity was not detected, increased connectivity between the striatal and brainstem/thalamus region to the sensorimotor network was observed [126]. Both task-based and rsfMRI have revealed alterations in brain activation levels and network recruitment following PAE. The changes in pattern and degree of network activation during tasks may indicate deficits in some networks typically used in healthy individuals, which forces individuals with FASD to use other resources to perform tasks to compensate for the deficiency. The use of both protocols in the same studies may help to elucidate FASD characteristics, given that individuals with FASD have a high rate of comorbid neurological impairments [127], which has confounded its diagnosis. 2.5.3. PET and SPECT PET and SPECT can be used to examine the neurotransmitter pathways and brain metabolism [16]. Both PET and SPECT have not been widely used in FASD research due to several shortcomings, including the requirement for injection of a radiotracer, long acquisition time, low spatial resolution and radiation concerns especially in infants and young children [16]. The number of participants in PET and SPECT studies was, therefore, not high compared to other modalities, and two of the SPECT studies were just case reports of individuals with FASD [23,128]. In a report on a single FASD-confirmed patient with facial dysmorphology, developmental disorders and a psychiatric history [128], in whom anatomical brain MRI had detected nothing abnormal, SPECT was the only modality that detected abnormalities. SPECT revealed decreased activity of Tc-99 m ethyl cysteinate dimer (a brain perfusion imaging agent) in the thalamus, basal ganglia and bilateral temporal lobes, indicating decreased blood flow in these regions [128]. A large reduction of blood flow in the temporal region relative to the cerebellum was observed in three subjects with FAS, while such reduction was smaller in two healthy subjects [23]. A SPECT study using Tc99 m hexamethylpropyleneamine oxime in children with FAS reported mild hypoperfusion in the left parietooccipital and right frontal region [48]. In another SPECT study using 123I–nor-β-CIT tracer, children with FASD were reported to have lower serotonin in the medial frontal cortex and higher levels of dopamine transporters in the striatum compared to healthy peers [129]. The only PET study, which used 18F–FDG, reported altered rate of glucose metabolism in the basal ganglia and thalamus in PAE children having normal IQ [130]. 2.6. Ultrasonography In studies of FASD, ultrasound is less widely used than other modalities discussed above. However, it offers the advantage of use in utero which is problematic in other modalities [131]. Two studies that used ultrasound imaging on infants with a confirmed history of PAE observed

V.T. Nguyen et al. / Magnetic Resonance Imaging 43 (2017) 10–26

malformation of the corpus callosum [132], and abnormality of the splenium [133], which is in good agreement with the MRI studies. Three other studies used ultrasound to monitor alcohol-exposed fetal development [134–136]. Alcohol-exposed fetuses showed smaller frontal cortex when scanned between the second and the third trimester [134]. Second trimester examination showed shorter caval-calvarial distance and frontothalamic distance, with the latter persisting into third trimester. During the third trimester, shorter orbital distance was also seen in the alcohol-exposed fetuses [136]. Compared to fetuses of abstaining mothers, fetuses developing under continuous alcohol exposure throughout gestation exhibited reduced cerebellar growth and ratio of cranial to body growth [135]. Given the wide availability of ultrasound and the ease with which it can be used for multiple follow-up imaging during development, the good agreement between ultrasound and other imaging modalities such as MRI suggests that ultrasound measures may be a useful tool for the early detection of PAE-induced abnormalities in utero and in infants, which can facilitate early implementation of appropriate interventions for infants with FASD [137]. 2.7. Insights from radiological studies of humans with FASD Of more than 100 studies discussed above, around one-fifth were performed prospectively while the remainder were carried out retrospectively. The majority of retrospective studies incorporated no information about the dosage, timing or duration of exposure, only a known history of maternal alcohol consumption reported by mothers or caregivers [138]. Another parameter not available in a number of studies is socioeconomic status (SES) which is known related to alcohol consumption patterns [139]. The concurrent consumption of alcohol and other drugs, such as marijuana or nicotine, is rather common [140], and while some studies (for example, see [27,33,41,48,71,74,79,82,84,85,96,114,120,124, 126,134,135]) considered maternal drug use as a covariate, other studies did not do so due to the lack of information. All these factors could have contributed to the heterogenous outcomes in imaging studies in humans with FASD. Nevertheless, studies in humans with FASD have provided valuable insight into the detrimental effects of PAE on the CNS. First, they have confirmed that brain abnormalities may result from not only heavy but also mild PAE (see studies that examined heavy and mild PAE separately, for example [5,49,59]), and also in the case of abstaining at some gestational point [47]. Vermis hypoplasia was observed in children born to mothers who had consumed alcohol in the first trimester, in the first two trimesters, as well as throughout gestation [47]. Second, they have suggested that the severity of impairments is dose dependent. The volume of alcohol intake per occasion was found inversely related to the cortical thickness in the right cuneus and pericalcarine cortices, the right occipitotemporal and parietal clusters [34]; and to volume reductions in broad medial regions [141]. The duration of exposure also mediated the magnitude of abnormalities. Moderate exposure throughout gestation resulted in significant changes in caudate volume, while changes resulted from first trimester exposure were not significantly different from no exposure [47,56]. Age at examination and the need for follow-up developmental studies are also the factors that should be taken into account in attempts to derive a consensus for CNS abnormalities in humans with FASD. Human brain development starts from the third week of gestation and continues postnatally, reaching approximately 90% of adult size at the age of 6, with structural maturation changes continuing until late adolescence [142]. Different parts of the brain have distinct development trajectories. For example, the human forebrain undergoes rapid growth until week 18 of gestation but continues to grow until 3 years after birth, while the cerebellum growth spurt starts about halfway through gestation and lasts until 18 months after birth [143]. The majority of FASD imaging studies recruited school-aged individuals (children and adolescents) (more than 70 studies), attributable to the fact that at this age children with ARND start to exhibit difficulties in learning,

17

which can be identified by their teachers [144]. Approximately 30 studies examined adolescents and young adults (11 to 30 years of age), 6 studies examined neonates and infants (less than 4 years of age) and 3 studies were carried out on fetuses. Only a few studies performed imaging at two time points to monitor developmental trajectories [21,26,31,44,51]. More follow-up studies are needed to investigate possible brain developmental repair and catch-up. CNS abnormalities revealed by the radiological imaging of humans with FASD are summarized in Table 1. 3. Radiological studies of translational animal models of PAE Studies in humans with FASD, even when obtained prospectively, cannot incorporate precise information about drinking quantities; moreover, virtually all drinking women drink repeatedly during pregnancy rather than at only a single time-point [138]. Animal models, on the other hand, have allowed specific investigations of the effects of PAE with respect to the quantity (from mild to heavy drinking), timing and duration of ethanol exposure during development, and demonstrated that the CNS exhibits varied vulnerability to ethanol insult across gestation [145]. The use of inbred strains and a controlled laboratory environment reduces potential confounders typically met in human studies, such as genetic heterogeneity, variable nutrition and substance abuse [146,147], and the involvement of alcohol dehydrogenase as a moderating variable [148]. Another advantage of animal models is that PAE-induced neuroanatomical changes that potentially underpin neurobehavioral changes can be further analyzed in tissue samples using molecular and histology methods [149], which are difficult to obtain in human studies. Animal models of PAE, therefore, offer great opportunity to provide additional insight into FASD in humans. Rodents are the most commonly used mammalian models for PAE studies, although nonhuman primates and sheep have also been used [150]. Rodent models offer advantages of lower cost, shorter gestational time (21 days in mice and rats compared to 23 to 26 weeks in nonhuman primates) [90] and ease of handling. The three weeks of gestation in mice corresponds approximately to the first and second trimester in humans in terms of brain development. Mouse brains continue to develop ten to twelve days postnatally, equivalent to the last trimester of human fetal brain development [151]. Therefore, mouse models designed to simulate third trimester PAE effects in humans requires exposures via lactation and/or directly on the neonates rather than through the placenta [150]. 3.1. Non-human primate PAE studies The similarities in brain development and neurobehaviours between non-human primates and humans make non-human primates the most suitable choice to study PAE. However, they have long gestation, are more expensive and difficult to handle compared to smaller animals such as rodents [150]. 3.1.1. An early non-human primate MRS study One of the earliest radiological studies involving an animal model of PAE was performed using chronic maternal ethanol exposure by oral administration during gestation in pigtailed macaques. Dose information was not provided, but the average maternal peak plasma ethanol concentration one week postpartum was 227 ± 33 mg/dl [90]. In the absence of detectable neuroanatomical changes in PAE group, MRS revealed significant increases in the Cho/Cr ratio in the thalamus, parts of the internal capsule and basal ganglia, and the adjacent white matter. The increased Cho/Cr ratio correlated with increased durations of PAE and increased neurobehavioral disorders in PAE animals. Increased Cho/Cr ratio has also been found in other neurological disorders such as multiple sclerosis and Alzheimer's disease [90]. Cho increases are commonly associated with cell destruction [152], but it can also be associated with demyelination [153]. This study showed that PAE-induced brain injury could be modeled in animals and observed non-invasively using MRS.

18

V.T. Nguyen et al. / Magnetic Resonance Imaging 43 (2017) 10–26

Table 1 Summary of radiological findings in humans with FASD. Techniques sMRI

DTI

Abnormalities Smaller brain size Gray Lobular volume Reduced volumes in: matter Total gray matter Frontal lobe Temporal lobe Parietal and occipital lobes. Parietal lobe more affected than temporal and occipital lobes. Reduced cortical volumes in: Left cuneus cortex, the lingual and right inferior temporal gyri Left parietooccipital and right anterior prefrontal cortices Left cingulate gyrus, the right middle frontal and temporal gyri Left superior frontal gyrus Middle frontal, supramarginal and inferior parietal regions. Increased cortical volumes in: Anterior and posterior cingulate regions and peri-Sylvian cortices. Lack of normal prolonged cortical volume increases. Abnormal linear volume increase in right and left transverse temporal regions. Regional volume Reduced regional volumes in: Cerebrum Cerebellum Cerebellar vermis Frontal lobe, caudate, putamen, hippocampus Thalamus Basal ganglia Globus pallidus, amygdala. Dysmorphic cerebellar vermis. Dysmorphic hippocampus and caudate nucleus. Cortical thickness Reduced cortical folding. and surface Reduced cortical surface in: Anterior cingulate cortex Right temporal lobe, right superior temporal gyrus and the region between the right temporal and occipital cortices Mid-sagittal anterior and posterior inferior lobes. Slower rate of cortical thinning. Reduced cortical thickness in: Globally Bilateral middle frontal lobe, central area, lateral inferior temporal and occipital lobes Left orbitofrontal region Occipitotemporal and parietal regions Bilateral superior parietal, right postcentral, left precentral and lateral occipital regions. Increased cortical thickness in: Lateral occipital regions Frontal, temporal and parietal regions. Density Increased gray matter density in: Parietal regions Unilateral peri-Sylvian cortices of temporal and parietal lobes. Frontal and occipital temporal regions. Wider cortical sulci. Lack of normal asymmetry in posterior temporal lobes. White matter Reduced white matter volumes in: Globally Occipital horns Left parietal and temporal regions Posterior cingulate region Ventral occipital temporal region Isthmus and splenium of corpus callosum Whole corpus callosum.

FA

Corpus callosum: Absent Thickened Displaced in the inferior–superior and anterior–posterior directions, more severely in the posterior part than anterior part. Cavum septi pellucidi and cavum vergae. Agenesis of hippocampal commissure Decreased in: Corpus callosum Splenium, posterior mid-body and isthmus Right cingulum, bilateral ILF and SLF and left thalamus Frontal, parietal and occipital lobes Medial prefrontal and posterior cingulate cortices Bilateral superior cerebellar peduncles Left middle peduncle.

References [5,17–23] [5,18,24,25] [5,20] [5,17,27] [17,27] [17,27] [28] [29] [30] [25] [31] [29] [26] [44] [45] [37,45–48] [20,49] [18,19,28–30,47,50,51] [18,29,51,52] [45,48,52–54] [18] [49] [55;56] [38] [36] [5] [37] [21] [18,21,22] [32] [33] [34] [35]

[39] [40] [41,42] [43] [42] [19,47,50] [27] [18,26] [48] [41,43] [57] [42] [59,64] [23,31,45,47,48,52, 58,60–63] [45] [61] [59] [58] [23] [70] [71–74] [74–76] [73,77] [78] [79,80] [80]

V.T. Nguyen et al. / Magnetic Resonance Imaging 43 (2017) 10–26

19

Table 1 (continued) Techniques RD

AD

MD

MRS

fMRI

NAA/Cr

NAA/Cho Cho/Cr Absolute values Working memory

Visual attention

Verbal memory function

Motor control Inhibition control Number processing function

rsfMRI

Abnormalities

References

Increased in bilateral globus pallidus. Increased in: Medial prefrontal and posterior cingulate cortices Superior cerebellar peduncles Left ILF and isthmus Left middle peduncle. Decreased in: Medial and inferior white matter Right SLF. Increased in: Bilateral inferior frontooccipital fasciculi, left ILF, SLF, corticospinal tracts, globus pallidus, right putamen, right thalamus. Left superior frontal lobe and anterior internal capsule Splenium Isthmus Whole corpus callosum Left middle and right inferior peduncles Cingulate, external capsule, uncinate fasciculus, superior corona radiate, arcuate fasciculus, frontal, fronto-central and central association fibers Cingulum bundles. Decreased in: Genu Right inferior frontal lobe and right temporo-parieto-occipital junction More steeply with age in the SLF and inferior frontooccipital fasciculi. Increased in (driven by increased NAA): Caudate nucleus. Decreased in (driven by decreased NAA): Frontal white matter, corpus callosum, thalamus and cerebellar dentate nucleus Right frontal/parietal white matter region and hippocampus. Decreased in parietal and frontal cortices (driven by decreased NAA). Decreased in right frontal/parietal white matter region and hippocampus (driven by decreased Cho). Decreased levels of NAA and Cho in deep cerebellar nuclei. 1-back task: Broader activation of cortical regions. N-back task, lower activation in: Right inferior frontal gyrus, right posterior parietal lobe, right dorsolateral prefrontal cortex and right middle frontal gyrus Superior frontal and parietal lobes Temporal lobe Basal ganglia Recruited a different subnetwork. From 0-back to 1-back task: Decreased activation in inferior frontal region Increased activation in basal ganglia and dorsal prefrontal cortex. Figure location identification: Higher activation levels relative to baseline condition in bilateral inferior, middle, and superior frontal gyri, superior temporal and occipital gyri, lentiform nuclei, and insulae. Higher activation of superior occipital temporal area. Higher activation of dorsal frontoparietal region in simple task and smaller increase in activation of frontal, parietal, temporal and lateral occipital regions from simple to more difficult task. Decreased functional activation in left medial temporal and dorsal prefrontal regions, left posterior temporal lobe, increased activation in left dorsal prefrontal cortex. Higher activation in left dorsal frontal and left inferior parietal cortices, and bilateral posterior temporal regions. Between rhythmic and non-rhythmic finger tapping task: Decreased activation difference in right crus I and vermis IV-V. Altered BOLD response in the fronto-striatal circuit. In arithmetic task: Lower deactivation of DMN Lower activation in left superior and right inferior parietal regions and medial frontal gyrus. In number proximity judgment: Lower activation in intraparietal sulcus Use a different network. In exact calculation, brain activation was more scattered and widespread. Decreased global deficiency. Lower functional connectivity between the medial prefrontal and posterior cingulate cortices. Lower interhemispheric functional connectivity in the para-central region.

[75]

Increased characteristic path length. Increased connectivity between striatal and brainstem/thalamus region to sensorimotor network.

[78] [79] [74] [79,80] [84] [85] [74–76] [73] [74,81] [74,82] [70,76] [79] [76] [78] [75] [73] [51] [93] [94] [95] [94] [95] [96] [103] [105] [106] [107] [108] [109] [109] [109] [110,111]

[42] [42,76] [112]

[113] [114] [115–117] [78] [118,119]

[120] [118] [121] [124] [78] [125] [124] [126] (continued on next page)

20

V.T. Nguyen et al. / Magnetic Resonance Imaging 43 (2017) 10–26

Table 1 (continued) Techniques PET and SPECT

SPECT

PET Ultrasono-graphy

Abnormalities

References

Decreased blood flow in thalamus, basal ganglia and bilateral temporal lobes. Hypoperfusion in left parietooccipital and right frontal regions. Large reduction of blood flow in the temporal region relative to the cerebellum Decreased serotonin in medial frontal cortex and increased dopamine transporters in the striatum. Altered rate of glucose metabolism in the basal ganglia and thalamus. Malformed corpus callosum. Smaller frontal cortex. Shorter caval-calvarial and frontothalamic distance, and shorter orbital distance. Reduced cerebellar growth and ratio cranial/body growth.

[128] [48] [23] [129] [130] [132,133] [134,136] [136] [135]

3.1.2. Non-human primate PET studies Only two PET studies of non-human primate models of PAE have been performed. These studies used rhesus macaques and focused on the brain dopamine system [154,155]. Dopamine is a neurotransmitter that promotes both excitatory and inhibitory responses. Dopamine plays an important role in motor function, attention and cognitive functions [155]. The density of postsynaptic dopamine receptors and the level of presynaptic dopamine synthesis are finely regulated to maintain their balance [154]. The macaque PAE studies used well-established radiotracers, 18F–Fallypride [154] and 11C–SCH23390 [155], to target the dopamine receptor D2 and D1, respectively. The macaques received a moderate concentration of ethanol (6% v/v consumed voluntarily daily, producing an average blood alcohol concentration of 20– 50 mg/dl 60 min after consumption), during (i) early gestation only (gestational day, GD, 0–50, equivalent to the first trimester in humans) [154], (ii) late gestation only (GD50–135) [154], or (iii) throughout gestation [154,155]. In [154], PAE in early gestation and throughout gestation resulted in a decrease of the dopamine D2 receptors to dopamine ratio in the striatum of young adult monkeys. Exposure during late gestation, however, caused the opposite effect. The absence of correlation between neonatal motor function and the dopamine D2 receptor to dopamine ratio in all PAE animals was believed to indicate a disruption of the striatal dopaminergic circuit [154]. In the second study, which used exposure throughout gestation, dopamine D1 receptor binding was found to be increased in the prefrontal cortex of PAE exposed male but not female adult offspring [155]. Findings from these macaque models of PAE suggest that the effects of ethanol differ depending on the gestational stage of exposure and that offspring gender may play a role in dopamine receptor vulnerability to PAE. The largest effect on the striatal dopaminergic circuit was observed in the group exposed throughout the gestation indicating that PAE-induced abnormalities are also dependent on the duration of exposure [154].

3.2. Other animal models of PAE Rodent models have been used to study the widest range of PAE patterns. A number of rodent studies have focused on the timing of exposure in order to investigate the stage dependence of ethanol teratogenesis. Gestation in mice can be grouped into: pre- and peri-implantation from GD1–6 and post-implantation (starting on GD6) including gastrulation and neurulation. Gastrulation and neurulation are critical time points, when neural plate development begins. Gastrulation usually begins at GD7, corresponding to 3 weeks 3 days to 4 weeks in humans, and neurulation begins on GD8 (week 4 in humans) and continues into GD10 [151]. 3.2.1. Mouse models of acute high dose of PAE In order to confirm the adverse effects of ethanol exposure on embryonic development, a series of studies using C57BL/6 J mice have used acute high doses of ethanol (from 23 to 25% v/v, injected twice intraperitoneally 4 h apart) at the following time-points: GD7, GD8, GD8.5, GD9, GD10 [151,156–159]. In all cases, morphology was examined ex vivo using MRI at GD17. 3.2.1.1. sMRI. Acute high dose ethanol exposure on a single day resulted in facial abnormalities and brain structural dysmorphology, however the defects were different for each time-point. GD7 ethanol exposure [151] resulted in midface abnormalities including long upper lip, closely spaced nostrils, narrow and pointed chin. MRI revealed olfactory bulb deficiency, frontal hemisphere fusion, lateral ventricle enlargement and malformation, third ventricle enlargement, short nasal septum and small nasal cavity, absence of the pituitary and septal region (Fig. 6), fusion of the striatal tissue, absence of corpus callosum, and presence of cortical heterotopia (Fig. 7). These findings are consistent with a later study that observed severe midfacial hypoplasia, shortening of palpebral fissures, elongated upper lip and

Fig. 6. Brain structural changes in acute, high dose ethanol-exposed mouse fetuses. MRI images show the absence of the septal region (long white arrow) in an ethanol-exposed mouse fetus at GD7 (b) compared to a control (a). The ventricles in the affected fetus (star) showed a significant enlargement and rostro-medial fusion. Adapted from [151].

V.T. Nguyen et al. / Magnetic Resonance Imaging 43 (2017) 10–26

21

Fig. 7. Cortical heterotopias in acute, high dose ethanol-exposed mouse fetuses. 3D reconstructed image shows heterotopias on the cortex (white arrows) (a) confirmed with histology (b) in C57Bl/6 J mouse fetus following acute 25% v/v ethanol exposure on GD7. Adapted from [151].

deficient philtrum in GD7 ethanol-exposed C57BL/6 J mice [158] and are consistent with FAS facial characteristics in humans [3]. GD8 ethanol exposure [157] caused a 25% reduction in total body volume and 19.5% reduction in brain volume. Overall brain volume reduction was traced back to many sub-regions, such as the cortex, striatum, hippocampus, cerebellum, and pons/medulla, but not to the septal region and the pituitary as was observed with GD7 exposure. The reduction in PAE brain regions co-occurred with larger ventricles, which has also been observed in humans [52]. The olfactory bulbs were poorly formed and widely spaced in the PAE mice. GD8.5 ethanol exposure resulted in inconsistent outcomes to exposure on GD7, with milder facial hypoplasia and palpebral fissure shortening, a shortened upper lip and preserved philtrum [158]. However, exposures at both GD7 and GD8.5 caused volumetric and shape abnormalities of the septal region, pituitary and olfactory bulbs, in contrast to the preserved septal and pituitary regions at GD8 [158]. GD9 ethanol exposure resulted in increased septal region width (9% wider) compared to non-exposed fetuses. The third ventricle of PAE animals also showed increased variability across the group, and the lateral, third and fourth ventricles were significantly larger in volume compared to controls. The cerebellum was 14% smaller with a significant change in shape, and bilaterally asymmetric, while the right and left cortices exhibited a trend towards slightly decreased volume. The cerebral cortex, hippocampus and right striatum were significantly distorted [159]. GD10 ethanol exposure produced similar outcomes to those of GD8 exposure, including decreased body size, and regional and total brain

volume reductions, except for larger third and lateral ventricles [156]. Heterotopias were also present in the GD10 ethanol-exposed mice, not on the cerebral cortical surface as on the GD7 but in the hypothalamic region, which intruded into the mildly enlarged third ventricle. The formation of heterotopias is believed to be a consequence of errors in cell migration, which have been linked to seizures and epilepsy [160], disorders with high prevalence in individuals with PAE [127,161,162].

3.2.1.2. Quantitative susceptibility mapping (QSM). QSM is a MRI technique that measures tissue magnetic property [163]. The only QSM study in rodents examined white matter tracts of ex vivo brains at postnatal day (PD) 45 in the offspring of C57Bl/6 J mice that were intraperitoneally injected with acute high doses of ethanol at GD7 (same model as in [151]) [163]. Following PAE, the volume of the anterior commissure was reduced, and three white matter regions including the corpus callosum, anterior commissure and hippocampal commissure showed lower susceptibility contrast with the surrounding gray matter compared to control mice. The reduction in susceptibility contrast between gray and white matter was due to an increase in positive magnetic susceptibility of white matter (i.e. becoming more paramagnetic), which may indicate demyelination and/or reduction in myelin formation in these regions. This study suggests that there may be other sensitive markers for PAE-induced microscopic changes, such as changes in susceptibility contrast, apart from the DTI-derived parameters which have been used in humans.

Fig. 8. 3D rendering skulls of PD28 mice by ex vivo CT. A healthy animal (a), leftward deviation of the frontal bone in a PAE animal (b) and smaller centroid area in another PAE animal (white arrow) (c). Adapted from [167].

22

V.T. Nguyen et al. / Magnetic Resonance Imaging 43 (2017) 10–26

3.2.2. Animal models of chronic high dose PAE 3.2.2.1. sMRI. Chronic high dose maternal ethanol exposure (25% w/v administered via intragastric gavage daily throughout gestation) in Long-Evans rats resulted in reduced brain volume, reduced neocortical volume and thickness, and altered neocortical surface in PAE offspring as compared to controls [164], analogous to findings in humans with FASD [5,32]. There were variations in the local cortical thickness, suggesting unequal vulnerabilities to ethanol in different regions of the cerebral cortex [164]. Ex vivo DTI revealed higher FA in PAE animals in the cerebral cortex, with strongest effect in the superficial layers throughout age range from PD0–6, proposedly due to less complexity of axonal and dendritic organization [165]. 3.2.2.2. CT imaging. A sheep model of chronic heavy PAE (40% w/v administered intravenously on three consecutive days a week) during the GD4– 41, equivalent to the first 8 weeks of human gestation, was examined at 6 months of age, a time-point equivalent to early adolescence in humans when individuals with FASD typically exhibited difficulties in learning [144]. CT imaging showed reductions in local skull bone volumes (frontal, parietal, occipital, maxilla, mandible, lacrimal, jugal bones) and total skull volume, decreased cranial circumference and decreased interorbital distance in peripubertal offspring [144]. These changes are similar to those reported in mouse models of PAE and humans with FASD (see above).

3.2.3. Rodent models of chronic moderate dose of PAE Ex vivo MRI at PD60 following PAE (10% v/v) throughout gestation revealed both increases and decreases in regional brain volumes in the offspring of 129Svev and C57Bl/6 mice, including decreased volumes of the olfactory bulbs, the granule cell layer of the dentate gyrus and the fourth ventricle; and increased volumes of the basal forebrain, the posterior part of the anterior commissure, the stria medullaris of the thalamus and the amygdala [166]. Moderate PAE (10% v/v provided ad libitum) from GD0–8 in C57Bl/ 6 J mice, encompassing pre-implantation, implantation and the first two days of gastrulation, equivalent to the first 3–4 weeks of human gestation [167], resulted in growth retardation (indicated by body weight) observable at the weaning (PD21) and extended into adulthood [167]. CT imaging revealed alterations in skull shape at PD28, including leftward deviation of the frontal bone (observed in one specimen), a smaller centroid area (Fig. 8), wider inter-orbital distance and shorter midface [167]. At PD60, ex vivo MRI revealed asymmetric changes in the hippocampus and olfactory bulb, in which significant increase in hippocampal volume and decrease in olfactory bulb volume were both observed in the left hemisphere [168]. 3.2.4. Rodent models of chronic low dose of PAE Ex vivo MRI at GD17 following low dose PAE (4.8% v/v) from GD7–11 in C57Bl/6 J mice, equivalent to weeks 3–6 in human gestation, resulted

Table 2 Summary of radiological findings in animal models of PAE. Technique Species MRS, PET

Dosage

Non-human Not provided primates (once a week, via oral tube) Moderate (6% v/v), daily, voluntarily

sMRI, QSM

Mice

Acute high dose (25% v/v) on a single day (two injections at 4 h interval)

Exposure stage

Abnormalities

References

Week 1–4, week 1–6 and throughout 24 weeks GD1–50 and across gestation

Increased Cho/Cr in thalamus, parts of internal capsules and basal ganglia with increase duration of exposure.

[90]

Altered ratios of dopamine D2 receptors to dopamine synthesis in striatum. Increased dopamine D1 receptor binding in prefrontal cortex.

[154]

GD7

Deficient midfacial deficiencies, olfactory bulbs. Frontal hemisphere fusion, ventricle enlargement. Abnormal nasal and septal region, striatal tissue. Smaller corpus callosum, cortical heterotopias. Reduced volume of anterior commissure. Reduced susceptibility contrast in corpus callosum, anterior commissure and hippocampal commissure. Reduced brain volume, deficient olfactory bulbs. Abnormal pituitary and septal regions. Shortened upper lip, malformed septal region, pituitary and olfactory bulbs. Wider septal region, larger ventricles. Smaller and asymmetric cerebellum. Deformed cortex, hippocampus and right striatum. Reduced brain volume, enlarged ventricles. Hypothalamic heterotopias. Decreased volumes of olfactory bulbs, granule cell layer of the dentate gyrus and fourth ventricle. Increased volumes of basal forebrain, posterior part of anterior commissure, stria medullaris of thalamus, and amygdala. Increased hippocampal volume and decreased olfactory bulb volume in the left hemisphere Skull dysmorphology. Growth retardation. Reduced parietal bone volume at PD7, Reduced frontal, parietal, occipital and mandible bone volumes at PD21, Reduced total skull volume, Reduced head circumference. Reduced volume of the cerebellum and increased volume of the septal region Reduced volume of the cerebellum and increased volume of the septal region Reduced brain volume, neocortical volume and thickness. Higher FA in cerebral cortex, with strongest effect in the superficial layers Reduced functional connectivity between primary and secondary visual cortex and other cortical components, between cerebellum and midbrain, and in inferior colliculi. Increased connectivity between cerebellum and striatal and cortical regions. Reduced local and total skull bone volumes, Reduced cranial circumference and interorbital distance.

GD8 GD8.5 GD9

GD10 Chronic moderate dose (10% v/v), daily, voluntarily

Across gestation

Chronic moderate dose (10% v/v), daily voluntarily

GD0–8

CT

MRI sMRI, DTI

Rats

rsfMRI

CT

Sheep

Chronic low dose (4.2% v/v), daily voluntarily

GD7–16

Chronic low dose (4.8% v/v), daily, voluntarily Chronic high dose (25% w/v), daily injections Chronic low dose (5% v/v), daily voluntarily

GD7–11 GD12–16 GD1–20

Chronic high dose (40% w/v), intravenous injections, 3 days/week

Across gestation

GD4–41

[155] [151,158]

[163]

[157] [158] [159]

[156] [166]

[168] [167] [173]

[169] [164] [165] [170]

[144]

V.T. Nguyen et al. / Magnetic Resonance Imaging 43 (2017) 10–26

in reduced volume of the cerebellum and increased volume of the septal region, while exposure from GD12–16, equivalent to weeks 7 to 12 in human gestation, resulted in reduced volume of the right hippocampus, and the pituitary gland [169]. Resting state fMRI was used to measure brain function in a model of low dose PAE (5% v/v) throughout gestation in Long-Evans rats [170]. Reductions were observed in the functional connectivity of the primary and secondary visual cortex to other cortical components, the cerebellum to the midbrain, and in the inferior colliculi; whereas increases in connectivity were observed in the cerebellum to the striatal and cortical regions. Notably, deficits in visual responses (mediated by visual cortex) and auditory responses (mediated by inferior colliculi) have been reported in children and young adults with FASD [171,172]. Low dose PAE (4.2% v/v) given 2 weeks prior to conception followed by exposure from GD7–16 in C57Bl/6 J mice resulted in global and regional reductions of head bone volumes [173]. CT imaging revealed reduced parietal bone volumes at PD7 and reduced frontal, parietal, occipital and mandible volumes at PD21, while total skull volume was decreased at both time-points. Head size and circumference were also decreased at both time-points. The smaller head size (microcephaly) and head circumference are consistent with features observed in humans with FAS [24,37]. 3.2.5. Insights from radiological studies of animal models of PAE Rodent models have been most widely used in FASD research, although unlike the sheep and non-human primates with long gestation and all three human-equivalent trimesters occurring in utero [174], in rodents, prenatally ethanol-exposed offspring cannot directly model third-trimester-equivalent alcohol-exposed individuals in humans [150]. In rodent models, in spite of different dosage, timing and duration, some structures appeared to have high susceptible to PAE, including the cerebellum, olfactory bulbs, hippocampus, striatum, corpus callosum, pituitary and septal regions. Animal models of PAE have demonstrated that PAE affects both brain and skull formation. This may be because both structures originate from the embryonic neural crest and undergo interdependent development [175]. Further studies are required to determine the specific developmental processes involved and the relationship between skull and brain malformations following PAE. Brain structural and functional abnormalities detected by radiological studies in animal models of PAE are summarized in Table 2. 3.3. Comparison of PAE effects in humans and animal models The literature on humans with FASD and animal models of PAE report a diverse range of outcomes, reflecting the wide spectrum of effects of alcohol exposure in utero. Both human and animal studies have shown that the pattern of PAE-induced brain abnormalities depends to some extent on the window of exposure during pregnancy. Nonetheless, similar structural changes have been observed, suggesting that some brain structures/subregions may exhibit higher susceptibilities to PAE than the others. Both animal models of PAE and studies of humans with FASD reported smaller brain size [17–20,95,157], smaller cerebellum [20,45–48,159], absence of the corpus callosum [45,151] or deformed hippocampus and caudate nucleus [55,159]. The abnormal growth and reductions in brain volume following PAE observed in humans and animals are believed to be a consequence of both apoptosis and incorrect cell migration. Both of these processes have been confirmed in a number of PAE studies using a histological approach (for example [176]). Asymmetric alterations within a single brain structure, such as the cerebellum in the mice following GD9 exposure [159] and the cortical lobes in humans [5,17,27] were also reported. MRS revealed metabolite level alterations in the caudate nucleus, frontal white matter and thalamus in both humans following moderate to high dose PAE [93] and pigtailed macaques following moderate chronic dose PAE [90]. Single high dose PAE on GD7 [151] and chronic

23

moderate dose PAE from GD0–8 in the mice [167] resulted in facial anomalies analogous to FAS facial characteristics in humans. Large scale changes, such as reduced physical growth and microcephaly in humans with FAS [5,17–23], were also reported in animal models of PAE [157,167,173]. Although animal models revealed facial characteristics analogous to FAS facial phenotype in humans, facial anomalies were not consistently observed in animals. The heavy drinking pattern across gestation is quite common in humans [177,178]. One of the few imaging studies using high dose PAE throughout gestation in Long-Evans rats produced good agreement with findings in humans, showing the features of reduced brain volume, neocortical volume and thickness, and altered cortical surface [18,36,38, 164]. However, the high dose regime at a single gestational day in animals showed consistent effect on the septal region in the mice [151,157–159], which has not been reported in humans with FASD. This may be attributable to the fact that a single high dose consumption during gestation is rarely the case in humans. There is also a gap in the literature for brain functional studies in animal models of PAE. Most brain receptor studies of PAE have been done using histology or in vitro ligand binding rather than in vivo [179]. RsfMRI is more advantageous compared to task-based fMRI since it is more applicable in animal studies. However, there is only one rsfMRI study in animals [170], which showed alterations in brain networks that mediate cognitive activities, including visual and auditory responses that had been found to be altered in humans with FASD [171, 172]. Such agreement, given the increasing availability of animal models, suggests that functional imaging like rsfMRI should be used more considerably in translational models of PAE to help answer questions about FASD phenotypes in humans. 4. Summary Modern radiological techniques have revealed numerous neurological changes which could underpin the neurobehavioral changes observed in individuals with FASD. While advanced imaging techniques offer many advantages, no single technique can provide a comprehensive insight into the underlying pathway of alcohol teratogenicity from the molecular level to macroscopic manifestation. There are notably gaps in the applications of radiology and MRI modalities in FASD research with limited use of CT imaging in humans, as is ultrasonography, diffusion MRI and rsfMRI in animals. Various animal models of PAE have been developed to mimic distinct PAE patterns. They have focused on certain brain structures and functions and enabled dissection of specific effects of PAE in high detail. The findings in FASD studies in humans and animal models of PAE are currently quite heterogenous yet convergent to some extent. Animal models, offering good controls over confounders such as substance abuse or SES, and the versatility in exposure patterns, therefore, play an important role in FASD research. They can provide opportunities to answer questions about the teratogenesis of alcohol in humans. A more systematic approach to radiological imaging in both humans and animal models, where multiple modalities and follow-up imaging are applied, is recommended. Such an approach would increase knowledge, aid diagnosis, enable the development and monitoring of suitable interventions for individuals with FASD. Funding This work was supported by National Health and Medical Research Council, Australia (grant number APP1003038). Van T Nguyen was supported by AUSAID scholarship. Acknowledgement We would like to thank Dr. Shona Osborne at the Centre for Advanced Imaging, The University of Queensland, Australia, for critical reading of the manuscript.

24

V.T. Nguyen et al. / Magnetic Resonance Imaging 43 (2017) 10–26

References [1] Jones KL, Smith DW. Recognition of the fetal alcohol sydrome in early pregnancy. The Lancet 1973:3. [2] Warren KR, Foudin LL. Alcohol-related birth defects - the past, present and future. Alcohol Res Health 2001;25(3):7. [3] Clarren SK. Recognition of Fetal Alcohol Sydrome. JAMA 1981;245(23):4. [4] Bertrand J, Floyd RL, Weber MK. National task force on fetal alcohol syndrome and fetal alcohol effect. Fetal alcohol sydrome: guidelines for referral and diagnosis; 2004. p. 62. [5] Rajaprakash M, et al. Cortical morphology in children with alcohol-related neurodevelopmental disorder. Brain Behav 2014;4(1):41–50. [6] Hoyme HE, et al. A practical clinical approach to diagnosis of fetal alcohol spectrum disorders: clarification of the 1996 institute of medicine criteria. Pediatrics 2005; 115(1):39–47. [7] Jacobson JL, Jacobson SW. Effects of prenatal alcohol exposure on child development. Alcohol Res Health 2002;26(4):5. [8] Riley EP, Infante MA, Warren KR. Fetal alcohol spectrum disorders: an overview. Neuropsychol Rev 2011;21(2):73–80. [9] Spadoni AD, et al. Neuroimaging and fetal alcohol spectrum disorders. Neurosci Biobehav Rev 2007;31(2):239–45. [10] Roozen S, et al. Worldwide prevalence of fetal alcohol spectrum disorders: a systematic literature review including meta-analysis. Alcohol Clin Exp Res 2016; 40(1):18–32. [11] Hanson JW, Jones KL, Smith DW. Fetal alcohol sydrome. experience with 41 patients. JAMA 1976;235(14):3. [12] Carter RC, et al. Fetal alcohol growth restriction and cognitive impairment. Pediatrics 2016;138(2). [13] Jaffer Z, Nelson M, Beighton P. Bone fusion in the fetal alcohol syndrome. J Bone Joint Surg 1981;63-B(4):3. [14] Gleason CA. Fetal alcohol exposure: Effects on the developing brain. NeoReviews 2001;2(10):7. [15] Clarren SK, et al. Brain malformatons related to prenatal exposure to ethanol. J Pediatr 1978;92(1):4. [16] Weiner M, et al. The use of MRI and PET for clinical diagnosis of dementia and investigation of cognitive impairment: a consensus report. Chicago, IL: Alzheimer's Association; 2005. [17] Archibald SL, et al. Brain dysmorphology in individuals with severe prenatal alcohol exposure. Dev Med Child Neurol 2001;43(03):148. [18] Nardelli A, et al. Extensive deep gray matter volume reductions in children and adolescents with fetal alcohol spectrum disorders. Alcohol Clin Exp Res 2011;35(8): 1404–17. [19] Coles CD, et al. Memory and brain volume in adults prenatally exposed to alcohol. Brain Cogn 2011;75(1):67–77. [20] Astley SJ, et al. Magnetic resonance imaging outcomes from a comprehensive magnetic resonance study of children with fetal alcohol spectrum disorders. Alcohol Clin Exp Res 2009;33(10):1671–89. [21] Treit S, et al. Longitudinal MRI reveals impaired cortical thinning in children and adolescents prenatally exposed to alcohol. Hum Brain Mapp 2014;35(9): 4892–903. [22] Treit S, et al. Relationships between head circumference, brain volume and cognition in children with prenatal alcohol exposure. PLoS One 2016;11(2):e0150370. [23] Bhatara VS, et al. Brain function in fetal alcohol syndrome assessed by single photon emission computed tomography. S D J Med 2002;55(2):4. [24] Rivkin MJ, et al. Volumetric MRI study of brain in children with intrauterine exposure to cocaine, alcohol, tobacco, and marijuana. Pediatrics 2008;121(4):741–50. [25] Soh DW, et al. Self-regulation therapy increases frontal gray matter in children with fetal alcohol spectrum disorder: evaluation by voxel-based morphometry. Front Hum Neurosci 2015;9:108. [26] Lebel C, et al. A longitudinal study of the long-term consequences of drinking during pregnancy: heavy in utero alcohol exposure disrupts the normal processes of brain development. J Neurosci 2012;32(44):15243–51. [27] Sowell EP, et al. Mapping cortical gray matter asymmetry patterns in adolescents with heavy prenatal alcohol exposure. Neuroimage 2002;17(4):1807–19. [28] Chen X, et al. Understanding specific effects of prenatal alcohol exposure on brain structure in young adults. Hum Brain Mapp 2012;33(7):1663–76. [29] Sowell ER, et al. Differentiating prenatal exposure to methamphetamine and alcohol versus alcohol and not methamphetamine using tensor-based brain morphometry and discriminant analysis. J Neurosci 2010;30(11):3876–85. [30] Eckstrand KL, et al. Persistent dose-dependent changes in brain structure in young adults with low-to-moderate alcohol exposure in utero. Alcohol Clin Exp Res 2012; 36(11):1892–902. [31] Gautam P, et al. Effects of prenatal alcohol exposure on the development of white matter volume and change in executive function. Neuroimage Clin 2014;5:19–27. [32] Zhou D, et al. Developmental cortical thinning in fetal alcohol spectrum disorders. Neuroimage 2011;58(1):16–25. [33] Gautam P, et al. Executive function and cortical thickness in youths prenatally exposed to cocaine, alcohol and tobacco. Dev Cogn Neurosci 2015;16:155–65. [34] Robertson FC, et al. Prenatal alcohol exposure is associated with regionally thinner cortex during the preadolescent period. Cereb Cortex 2016;26(7):3083–95. [35] Infante MA, et al. Atypical cortical gyrification in adolescents with histories of heavy prenatal alcohol exposure. Brain Res 2015;1624:9. [36] Migliorini R, et al. Anterior cingulate cortex surface area relates to behavioral inhibition in adolescents with and without heavy prenatal alcohol exposure. Behav Brain Res 2015;292:10.

[37] Cardenas VA, et al. Automated cerebellar segmentation: validation and application to detect smaller volumes in children prenatally exposed to alcohol. Neuroimage Clin 2014;4:295–301. [38] De Guio F, et al. A study of cortical morphology in children with fetal alcohol spectrum disorders. Hum Brain Mapp 2014;35(5):2285–96. [39] Sowell ER, et al. Abnormal cortical thickness and brain-behavior correlation patterns in individuals with heavy prenatal alcohol exposure. Cereb Cortex 2008;18(1):136–44. [40] Yang Y, et al. Abnormal cortical thickness alterations in fetal alcohol spectrum disorders and their relationships with facial dysmorphology. Cereb Cortex 2012; 22(5):1170–9. [41] Sowell ER, et al. Regional brain shape abnormalities persist into adolescence after heavy prenatal alcohol exposure. Cereb Cortex 2002;12(8):9. [42] Li Z, et al. Occipital-temporal reduction and sustained visual attention deficit in prenatal alcohol exposed adults. Brain Imaging Behav 2008;2(1):39–48. [43] Sowell EP, et al. Voxel-based morphometric analyses of the brain in children and adolescents prenatally exposed to alcohol. Cogn Neurosci Neuropsychol 2001;12(3):9. [44] Gautam P, et al. Volume changes and brain-behavior relationships in white matter and subcortical gray matter in children with prenatal alcohol exposure. Hum Brain Mapp 2015;36(6):2318–29. [45] Mattson SN, Jernigan TL, Riley EP. MRI and prenatal alcohol exposure. Alcohol Health Res World 1994;18(1):4. [46] Sowell ER, et al. Abnormal development of the cerebellar vermis in children prenatally exposed to alcohol: size reduction in lobules I\ \V. Alcohol Clin Exp Res 1996;20(1):4. [47] Autti-Rämö I, et al. MRI findings in children with school problems who had been exposed prenatally to alcohol. Dev Med Child Neurol 2002;44(02):98. [48] Riikonen R, et al. Brain perfusion SPECT and MRI in fetal alcohol syndrome. Dev Med Child Neurol 1999;41:8. [49] O'Hare ED, et al. Mapping cerebellar vermal morphology and cognitive correlates in prenatal alcohol exposure. Dev Neurosci 2005;16(12):6. [50] Willoughby KA, et al. Effects of prenatal alcohol exposure on hippocampal volume, verbal learning, and verbal and spatial recall in late childhood. J Int Neuropsychol Soc 2008;14(6):12. [51] Treit S, et al. Longitudinal MRI reveals altered trajectory of brain development during childhood and adolescence in fetal alcohol spectrum disorders. J Neurosci 2013; 33(24):10098–109. [52] Mattson SN, et al. Fetal Alcohol syndrome: a case report of neuropsychological, MRI, and EEG assessment of two children. Alcohol Clin Exp Res 1992;16(3):3. [53] Mattson SN, et al. A decrease in the size of the basal ganglia in children with fetal alcohol syndrome. Alcohol Clin Exp Res 1996;20(6):6. [54] Roussotte FF, et al. Regional brain volume reductions relate to facial dysmorphology and neurocognitive function in fetal alcohol spectrum disorders. Hum Brain Mapp 2012;33(4):920–37. [55] Joseph J, et al. Three-dimensional surface deformation-based shape analysis of hippocampus and caudate nucleus in children with fetal alcohol spectrum disorders. Hum Brain Mapp 2014;35(2):659–72. [56] Willford J, et al. Caudate asymmetry: a neurobiological marker of moderate prenatal alcohol exposure in young adults. Neurotoxicol Teratol 2010;32(6):589–94. [57] Bjorkquist OA, et al. Cingulate gyrus morphology in children and adolescents with fetal alcohol spectrum disorders. Psychiatry Res 2010;181(2):101–7. [58] Swayze VW, et al. Magnetic resonance imaging of brain anomalies in fetal alcohol syndrome. Pediatrics 1997;99(2):11. [59] Sowell ER, et al. Mapping callosal morphology and cognitive correlates: effects of heavy prenatal alcohol exposure. Neurology 2001;57(2):10. [60] Riley EP, et al. Abnormalities of the corpus callosum in children prenatally exposed to alcohol. Alcohol Clin Exp Res 1995;19(5):5. [61] Bookstein FL, et al. Corpus callosum shape and neuropsychological deficits in adult males with heavy fetal alcohol exposure. Neuroimage 2002;15(1):233–51. [62] Bookstein FL, et al. Midline corpus callosum is a neuroanatomical focus of fetal alcohol damage. Anat Rec 2002;269(3):162–74. [63] Yang Y, et al. Callosal thickness reductions relate to facial dysmorphology in fetal alcohol spectrum disorders. Alcohol Clin Exp Res 2012;36(5):798–806. [64] Dodge NC, et al. Prenatal Alcohol Exposure and Interhemispheric Transfer of Tactile Information: Detroit and Cape Town Findings. Alcohol Clin Exp Res 2009;33(9):10. [65] Grahn JA, Parkinson JA, Owen AM. The cognitive functions of the caudate nucleus. Prog Neurobiol 2008;86(3):141–55. [66] Fryer SL, et al. Caudate volume predicts neurocognitive performance in youth with heavy prenatal alcohol exposure. Alcohol Clin Exp Res 2012;36(11):1932–41. [67] Roebuck TM, Mattson SN, Riley EP. Interhemispheric transfer in children with heavy prenatal alcohol exposure. Alcohol Clin Exp Res 2002;26(12):9. [68] Bihan DL, et al. Diffusion tensor imaging: concepts and applications. J Magn Reson Imaging 2001;13:13. [69] Alexander AL, et al. Diffusion Tensor Imaging of the brain. Neurotherapeutics 2007; 4:14. [70] Ma X, et al. Evaluation of corpus callosum anisotropy in young adults with fetal alcohol syndrome according to diffusion tensor imaging. Alcohol Clin Exp Res 2005; 29(7):1214–22. [71] Wozniak JR, et al. Microstructural corpus callosum anomalies in children with prenatal alcohol exposure: an extension of previous diffusion tensor imaging findings. Alcohol Clin Exp Res 2009;33(10):1825–35. [72] Fryer SL, et al. Characterization of white matter microstructure in fetal alcohol spectrum disorders. Alcohol Clin Exp Res 2009;33(3):514–21. [73] Li L, et al. Voxelwise and skeleton-based region of interest analysis of fetal alcohol syndrome and fetal alcohol spectrum disorders in young adults. Hum Brain Mapp 2009;30(10):3265–74. [74] Fan J, et al. White matter deficits mediate effects of prenatal alcohol exposure on cognitive development in childhood. Hum Brain Mapp 2016;37(8):16.

V.T. Nguyen et al. / Magnetic Resonance Imaging 43 (2017) 10–26 [75] Lebel C, et al. Brain diffusion abnormalities in children with fetal alcohol spectrum disorder. Alcohol Clin Exp Res 2008;32(10):1732–40. [76] O'Conaill CR, et al. Visual search for feature conjunctions: an fMRI study comparing alcohol-related neurodevelopmental disorder (ARND) to ADHD. J Neurodev Disord 2015;7(1):10. [77] Sowell ER, et al. Mapping white matter integrity and neurobehavioral correlates in children with fetal alcohol spectrum disorders. J Neurosci 2008;28(6):1313–9. [78] Santhanam P, et al. Default mode network dysfunction in adults with prenatal alcohol exposure. Psychiatry Res 2011;194(3):354–62. [79] Fan J, et al. White matter integrity of the cerebellar peduncles as a mediator of effects of prenatal alcohol exposure on eyeblink conditioning. Hum Brain Mapp 2015;36(7):13. [80] Spottiswoode BS, et al. Diffusion tensor imaging of the cerebellum and eyeblink conditioning in fetal alcohol spectrum disorder. Alcohol Clin Exp Res 2011; 35(12):2174–83. [81] Paolozza A, et al. Response inhibition deficits in children with fetal alcohol spectrum disorder: relationship between diffusion tensor imaging of the corpus callosum and eye movement control. Neuroimage Clin 2014;5:9. [82] Wozniak JR, et al. Diffusion tensor imaging in children with fetal alcohol spectrum disorders. Alcohol Clin Exp Res 2006;30(10):8. [83] Winston GP. The physical and biological basis of quantitative parameters derived from diffusion MRI. Quant Imaging Med Surg 2012;2(4):254–65. [84] Taylor PA, et al. A DTI-based tractography study of effects on brain structure associated with prenatal alcohol exposure in newborns. Hum Brain Mapp 2015;36(1): 170–86. [85] Donald KA, et al. A study of the effects of prenatal alcohol exposure on white matter microstructural integrity at birth. Acta Neuropsychiatr 2015;27(4):9. [86] Lebel C, et al. Brain microstructure is related to math ability in children with fetal alcohol spectrum disorder. Alcohol Clin Exp Res 2010;34(2):10. [87] Green CR, et al. Diffusion tensor imaging correlates of saccadic reaction time in children with fetal alcohol spectrum disorder. Alcohol Clin Exp Res 2013;37(9): 1499–507. [88] Schmitt LM, et al. Saccadic eye movement abnormalities in autism spectrum disorder indicate dysfunctions in cerebellum and brainstem. Mol Autism 2014;5(1):13. [89] Gujar SK, et al. Magnetic Resonance Spectroscopy. Neuro-Ophthalmol 2005;25(3): 10. [90] Astley SJ, et al. Magnetic resonance imaging and spectroscopy in fetal ethanol exposed Macaca nemestrina. Neurotoxicol Teratol 1995;17(5):8. [91] Moffett JR, et al. N-acetylaspartate in the CNS: from neurodiagnostics to neurobiology. Prog Neurobiol 2007;81(2):43. [92] Everitt B, Robbins TW. Central cholinergic systems and cognition. Annu Rev Psychol 1997;48:36. [93] Cortese BM, et al. Magnetic resonance and spectroscopic imaging in prenatal alcohol-exposed children: preliminary findings in the caudate nucleus. Neurotoxicol Teratol 2006;28(5):597–606. [94] Fagerlund A, et al. Brain metabolic alterations in adolescents and young adults with fetal alcohol spectrum disorders. Alcohol Clin Exp Res 2006;30(12):2097–104. [95] Astley SJ, et al. Magnetic resonance spectroscopy outcomes from a comprehensive magnetic resonance study of children with fetal alcohol spectrum disorders. Magn Reson Imaging 2009;27(6):760–78. [96] du Plessis L, et al. An in vivo 1H magnetic resonance spectroscopy study of the deep cerebellar nuclei in children with fetal alcohol spectrum disorders. Alcohol Clin Exp Res 2014;38(5):1330–8. [97] Matalon R, Michals-Matalon K. Canavan DiseaseGeneReviews ; 1999. [98] Glover GH. Overview of functional magnetic resonance imaging. Neurosurg Clin N Am 2011;22(2):133–9 vii. [99] Crosson B, et al. Functional imaging and related techniques: an introduction for rehabilitation researchers. J Rehabil Res Dev 2010;47(2):33. [100] Graham DM, et al. Prenatal alcohol exposure, attention-deficit/hyperactivity disorder, and sluggish cognitive tempo. Alcohol Clin Exp Res 2013;37(Suppl. 1): E338–46. [101] Infante MA, et al. Objective assessment of ADHD core symptoms in children with heavy prenatal alcohol exposure. Physiol Behav 2015;148:45–50. [102] Crocker N, et al. Comparison of adaptive behavior in children with heavy prenatal alcohol exposure or attention-deficit/hyperactivity disorder. Alcohol Clin Exp Res 2009;33(11):2015–23. [103] Malisza KL, et al. Comparison of spatial working memory in children with prenatal alcohol exposure and those diagnosed with ADHD; A functional magnetic resonance imaging study. J Neurodev Disord 2012;4(1):12. [104] Coles CD, et al. A comparison of children affected by prenatal alcohol exposure and attention deficit, hyperactity disorder. Alcohol Clin Exp Res 1997;21(1):12. [105] Astley SJ, et al. Functional magnetic resonance imaging outcomes from a comprehensive magnetic resonance study of children with fetal alcohol spectrum disorders. J Neurodev Disord 2009;1(1):61–80. [106] Malisza KL, et al. Evaluation of spatial working memory function in children and adults with fetal alcohol spectrum disorders: a functional magnetic resonance imaging study. Pediatr Res 2005;58(6):1150–7. [107] Gautam P, et al. Developmental trajectories for visuo-spatial attention are altered by prenatal alcohol exposure: a longitudinal fMRI study. Cereb Cortex 2015; 25(12):4761–71. [108] Roussotte FF, et al. Abnormal brain activation during working memory in children with prenatal exposure to drugs of abuse: the effects of methamphetamine, alcohol, and polydrug exposure. Neuroimage 2011;54(4):3067–75. [109] Diwadkar VA, et al. Differences in cortico-striatal-cerebellar activation during working memory in syndromal and nonsyndromal children with prenatal alcohol exposure. Hum Brain Mapp 2013;34(8):15.

25

[110] Spadoni AD, et al. BOLD response during spatial working memory in youth with heavy prenatal alcohol exposure. Alcohol Clin Exp Res 2009;33(12):2067–76. [111] Norman AL, et al. A functional magnetic resonance imaging study of spatial working memory in children with prenatal alcohol exposure: contribution of familial history of alcohol use disorders. Alcohol Clin Exp Res 2013;37(1):132–40. [112] Sowell ER, et al. Functional magnetic resonance imaging of verbal learning in children with heavy prenatal alcohol exposure. Neuroreport 2007;18(7):5. [113] O'Hare ED, et al. Altered frontal-parietal functioning during verbal working memory in children and adolescents with heavy prenatal alcohol exposure. Hum Brain Mapp 2009;30(10):3200–8. [114] du Plessis L, et al. Neural correlates of cerebellar-mediated timing during finger tapping in children with fetal alcohol spectrum disorders. Neuroimage Clin 2015; 7:562–70. [115] Fryer SL, et al. Prenatal alcohol exposure affects frontal-striatal BOLD response during inhibitory control. Alcohol Clin Exp Res 2007;31(8):1415–24. [116] O'Brien JW, et al. Effect of predictive cuing on response inhibition in children with heavy prenatal alcohol exposure. Alcohol Clin Exp Res 2013;37(4):644–54. [117] Ware AL, et al. An fMRI study of behavioral response inhibition in adolescents with and without histories of heavy prenatal alcohol exposure. Behav Brain Res 2015; 278:10. [118] Santhanam P, et al. Effects of prenatal alcohol exposure on brain activation during an arithmetic task: an fMRI study. Alcohol Clin Exp Res 2009;33(11):1901–8. [119] Lv J, et al. Assessing effects of prenatal alcohol exposure using group-wise sparse representation of fMRI data. Psychiatry Res 2015;233(2):254–68. [120] Woods KJ, et al. Parietal dysfunction during number processing in children with fetal alcohol spectrum disorders. Neuroimage Clin 2015;8:594–605. [121] Meintjes EM, et al. An FMRI study of number processing in children with fetal alcohol syndrome. Alcohol Clin Exp Res 2010;34(8):1450–64. [122] Lee MH, Smyser CD, Shimony JS. Resting-state fMRI: a review of methods and clinical applications. AJNR Am J Neuroradiol 2013;34(10):1866–72. [123] Koch W, et al. Effects of aging on default mode network activity in resting state fMRI: does the method of analysis matter? Neuroimage 2010;51(1):280–7. [124] Wozniak JR, et al. Global functional connectivity abnormalities in children with fetal alcohol spectrum disorders. Alcohol Clin Exp Res 2013;37(5):748–56. [125] Wozniak JR, et al. Inter-hemispheric functional connectivity disruption in children with prenatal alcohol exposure. Alcohol Clin Exp Res 2011;35(5):849–61. [126] Donald KA, et al. Interhemispheric functional brain connectivity in neonates with prenatal alcohol exposure: preliminary findings. Alcohol Clin Exp Res 2016; 40(1):9. [127] Burd L, et al. Recognition and management of fetal alcohol syndrome. Neurotoxicol Teratol 2003;25(6):681–8. [128] Codreanu I, Yang J, Zhuang H. Brain single-photon emission computed tomography in fetal alcohol syndrome: a case report and study implications. J Child Neurol 2012;27(12):1580–4. [129] Riikonen RS, et al. Deep serotonergic and dopaminergic structures in fetal alcoholic syndrome: a study with nor-beta-CIT-single-photon emission computed tomography and magnetic resonance imaging volumetry. Biol Psychiatry 2005;57(12): 1565–72. [130] Clark CM, et al. Structural and functional brain integrity of fetal alcohol syndrome in nonretarded cases. Pediatrics 2000;105(5):4. [131] Abramowicz JS. Benefits and risks of ultrasound in pregnancy. Semin Perinatol 2013;37(5):295–300. [132] Grant TM, et al. Neonatal cranial ultrasound leads to early diagnosis and intervention in baby of alcohol-abusing mother. Ment Health Aspects Dev Disabil 2006; 9(4):3. [133] Bookstein FL, et al. Preliminary evidence that prenatal alcohol damage may be visible in averaged ultrasound images of the neonatal human corpus callosum. Alcohol 2005;36(3):151–60. [134] Wass TS, Persutte WH, Hobbins JC. The impact of prenatal alcohol exposure on frontal cortex development in utero. Am J Obstet Gynecol 2001;185:6. [135] Handmaker NS, et al. Impact of alcohol exposure after pregnancy recognition on ultrasonographic fetal growth measures. Alcohol Clin Exp Res 2006;30(5):7. [136] Kfir M, et al. Can prenatal ultrasound detect the effects of in-utero alcohol exposure? A pilot study. Ultrasound Obstet Gynecol 2009;33(6):7. [137] Chudley AE, et al. Fetal alcohol spectrum disorder: Canadian guidelines for diagnosis. CMAJ 2005;172(5 Suppl):S1–S21. [138] Benz J, Rasmussen C, Andrew G. Diagnosing fetal alcohol spectrum disorder: History, challenges and future directions. Paediatr Child Health 2009;14(4):7. [139] Huckle T, You RQ, Casswell S. Socio-economic status predicts drinking patterns but not alcohol-related consequences independently. Addiction 2010;105(7): 1192–202. [140] Forray A. Substance use during pregnancy. , 5F1000Research; 2016 9. [141] Meintjes EM, et al. A tensor-based morphometry analysis of regional differences in brain volume in relation to prenatal alcohol exposure. Neuroimage Clin 2014;5:9. [142] Stiles J, Jernigan TL. The basics of brain development. Neuropsychol Rev 2010; 20(4):327–48. [143] Neligan GA. The human brain growth spurt. Dev Med Child Neurol 1974;16(5):2. [144] Birch SM, et al. Computed tomography assessment of peripubertal craniofacial morphology in a sheep model of binge alcohol drinking in the first trimester. Alcohol 2015;49(7):675–89. [145] O'Leary-Moore SK, et al. Focus on: magnetic resonance-based studies of fetal alcohol spectrum disorders in animal models. Alcohol Res Health 2011;34(1):7. [146] Wilson SE, Cudd TA. Focus on: the use of animal models for the study of fetal alcohol spectrum disorders. Alcohol Res Health 2011;34(1):7. [147] Bailey BA, Sokol RJ. Prenatal alcohol exposure and miscarriage, stillbirth, preterm delivery, and sudden infant death syndrome. Alcohol Res Health 2011;34(1):6.

26

V.T. Nguyen et al. / Magnetic Resonance Imaging 43 (2017) 10–26

[148] Warren KR, Li TK. Genetic polymorphisms: impact on the risk of fetal alcohol spectrum disorders. Birth Defects Res A Clin Mol Teratol 2005;73(4):195–203. [149] Toga AW, et al. Towards multimodal atlases of the human brain. Nat Rev Neurosci 2006;7(12):952–66. [150] Patten AR, Fontaine CJ, Christie BR. A comparison of the different animal models of fetal alcohol spectrum disorders and their use in studying complex behaviors. Front Pediatr 2014;2:93. [151] Godin EA, et al. Magnetic resonance microscopy defines ethanol-induced brain abnormalities in prenatal mice: effects of acute insult on gestational day 7. Alcohol Clin Exp Res 2010;34(1):98–111. [152] Pandya HG, et al. The nonspecific nature of proton spectroscopy in brain masses in children: a series of demyelinating lesions. Neuroradiology 2005;47(12):955–9. [153] Rovira A, Alonso J. 1H magnetic resonance spectroscopy in multiple sclerosis and related disorders. Neuroimaging Clin N Am 2013;23(3):459–74. [154] Schneider ML, et al. Moderate-level prenatal alcohol exposure alters striatal dopamine system function in rhesus monkeys. Alcohol Clin Exp Res 2005;29(9):1685–97. [155] Converse AK, et al. Moderate-level prenatal alcohol exposure induces sex differences in dopamine d1 receptor binding in adult rhesus monkeys. Alcohol Clin Exp Res 2014;38(12):2934–43. [156] O'Leary-Moore SK, et al. Magnetic resonance microscopy-based analyses of the brains of normal and ethanol-exposed fetal mice. Birth Defects Res A Clin Mol Teratol 2010;88(11):953–64. [157] Parnell SE, et al. Magnetic resonance microscopy defines ethanol-induced brain abnormalities in prenatal mice: effects of acute insult on gestational day 8. Alcohol Clin Exp Res 2009;33(6):1001–11. [158] Lipinski RJ, et al. Ethanol-induced face-brain dysmorphology patterns are correlative and exposure-stage dependent. PLoS One 2012;7(8):e43067. [159] Parnell SE, et al. Magnetic resonance microscopy-based analyses of the neuroanatomical effects of gestational day 9 ethanol exposure in mice. Neurotoxicol Teratol 2013;39:77–83. [160] Verrotti A, et al. New trends in neuronal migration disorders. Eur J Paediatr Neurol 2010;14(1):1–12. [161] Sun Y, et al. Binge drinking during pregnancy and risk of seizures in childhood: a study based on the Danish National Birth Cohort. Am J Epidemiol 2009;169(3): 313–22. [162] Bell SH, et al. The remarkably high prevalence of epilepsy and seizure history in fetal alcohol spectrum disorders. Alcohol Clin Exp Res 2010;34(6):9. [163] Cao W, et al. Prenatal alcohol exposure reduces magnetic susceptibility contrast and anisotropy in the white matter of mouse brains. Neuroimage 2014;102(Pt 2):748–55.

[164] Leigland LA, et al. The influence of fetal ethanol exposure on subsequent development of the cerebral cortex as revealed by magnetic resonance imaging. Alcohol Clin Exp Res 2013;37(6):924–32. [165] Leigland LA, et al. Diffusion MRI of the developing cerebral cortical gray matter can be used to detect abnormalities in tissue microstructure associated with fetal ethanol exposure. Neuroimage 2013;83:1081–7. [166] Akers KG, et al. Fetal alcohol exposure leads to abnormal olfactory bulb development and impaired odor discrimination in adult mice. Mol Brain 2011;4:29. [167] Kaminen-Ahola N, et al. Maternal ethanol consumption alters the epigenotype and the phenotype of offspring in a mouse model. PLoS Genet 2010;6(1):e1000811. [168] Marjonen H, et al. Early maternal alcohol consumption alters hippocampal DNA methylation, gene expression and volume in a mouse model. PLoS One 2015; 10(5):e0124931. [169] Parnell SE, et al. Dysmorphogenic effects of first trimester-equivalent ethanol exposure in mice: a magnetic resonance microscopy-based study. Alcohol Clin Exp Res 2014;38(7):2008–14. [170] Rodriguez CI, et al. Moderate prenatal alcohol exposure alters functional connectivity in the adult rat brain. Alcohol Clin Exp Res 2016;40(10):2134–46. [171] Coffman BA, et al. Primary visual response (M100) delays in adolescents with FASD as measured with MEG. Hum Brain Mapp 2013;34(11):2852–62. [172] Stephen JM, et al. Delays in auditory processing identified in preschool children with FASD. Alcohol Clin Exp Res 2012;36(10):1720–7. [173] Shen L, et al. Effect of prenatal alcohol exposure on bony craniofacial development: a mouse MicroCT study. Alcohol 2013;47(5):405–15. [174] Ramadoss J, et al. Maternal adrenocorticotropin, cortisol, and thyroid hormone responses to all three-trimester equivalent repeated binge alcohol exposure: ovine model. Alcohol 2008;42(3):199–205. [175] Richtsmeier JT, Flaherty K. Hand in glove: brain and skull in development and dysmorphogenesis. Acta Neuropathol 2013;125(4):469–89. [176] West JR, et al. Cell population depletion associated with fetal alcohol brain damage: mechanisms of BAC-dependent cell loss. Alcohol Clin Exp Res 1990;14(6):8. [177] May PA, et al. The epidemiology of fetal alcohol sydrome and partial FAS in a South African community. Drug Alcohol Depend 2007;88(2-3):13. [178] Shepherd SM, et al. Violence risk assessment in australian aboriginal offender populations: a review of the literature. Psychol Public Policy Law 2014;20(3). [179] Cherry SR, Gambhir SS. Use of Positron Emission Tomography in Animal Research. ILAR J 2001;42(3):14. [180] Warren KR, H.B.G., Thomas JD. Fetal alcohol spectrum disorders research challenges and opportunities. Alcohol Res Health 2011;34(1):11.