Embed Size (px)
Citation preview
P E R S P E C T I V E
www.ScienceTranslationalMedicine.org 16 January 2013 Vol 5 Issue 168 168ps2 1
Development of the brain’s cerebral cortex is regulated by a complex interplay between an unfolding genetic program and the en-vironment. Neuronal proliferation, migra-tion, dif erentiation, and circuit formation follow def ned time scales, and their cho-reography is controlled by extrinsic factors mediated through blood-brain and placen-tal barriers (1). Premature birth can alter the normal developmental processes. If the preterm brain does not possess the ability to recover from these disruptions, then they may be associated with neurodevelopmental impairments, with some consequences not manifesting until years later. Understand-ing the timing and rationale for therapeutic intervention to prevent developmental ab-normalities in these infants requires the de-tailed monitoring of clinical parameters and the study of mechanistic aspects in animals.
Two papers published in this issue of Sci-ence Translational Medicine characterize ad-verse neurodevelopmental outcomes in pre-term newborn infants. Vinall and colleagues (2) examine the association between neona-tal growth (weight, length, and head size) and dif usion tensor imaging (DTI) measures of cortical development in infants born preterm (between 24 and 32 weeks gestational ages). Dean and colleagues (3) use a sheep model to examine the consequences of prenatal ce-rebral ischemia on cortical volume and relate magnetic resonance imaging (MRI)–def ned cortical microstructure with brain histologi-cal analysis. Both of these translational stud-ies draw attention to grey matter damage and suggest that avoiding somatic growth impair-
ment during neonatal care may allow cortical development to proceed optimally and thus reduce neurological disabilities related to preterm birth.
DECIPHERING THE DEVELOPING CORTEXIt is well documented that survivors of pre-term birth show a reduction in cerebral cortical volume relative to infants born at term. Premature birth—even in the absence of overt hypoxic-ischemic injury—is associ-ated with loss of both cortical and subcorti-cal (thalamus and basal ganglia) grey-matter volume, and the more preterm the infant, the greater the reduction in volume (4). T is re-duction in the volumes of cortical and sub-cortical (thalamus, basal ganglia) grey mat-ter is not always associated with overt signs of white matter volume loss or injury. T ese observations are supported by the study by Vinall et al. (2), which showed that lower gestational age is associated with higher cor-tical fractional anisotropy (FA)—a dif usion imaging measure that ref ects axon density and diameter as well as extent of myelina-tion (white matter). FA values change during development of the cerebral cortex and are higher at early developmental stages because the majority of processes are radial—that is, structures mature and migrate from sites of neurogenesis (Fig. 1). Later, af er completion of neurogenesis, radial glia progenitors dis-appear or are transformed to astrocytes (star-shaped glial cells that support neurons of the brain and contribute to the blood-brain barrier); cortical connections that transport nerve impulses from sense organs (af erent) and to subcortical structures (ef erent) ma-ture; and neurons develop extensive branch-ing and arborization (treelike arrangement of processes). T ese processes are ref ected in a normal reduction in FA values with increas-ing age of the cortex.
T e Vinall et al. (2) study used a noninva-sive imaging method—dif usion tensor MRI [at 1.5 Tesla (T), a unit of measurement that
indicates the strength of a magnetic f eld]—to map the dif usion of water in the devel-oping brains of a large cohort (N = 95) of newborn preterm human infants (neonates). T e dif usion patterns of water provide an indirect measure of the existing macromo-lecular and structural elements—“obstacles” that alter f uid f ow in the organ being mea-sured, decreasing anisotropy—and thus yield information about the state (normal or abnormal) of the tissue architecture. T e authors investigated changes in the cortical microstructure of the preterm neonates over time; scan 1 was taken at ~32 weeks gesta-tion and scan 2 at ~40 weeks postmenstrual age (term equivalent). From these scans, Vi-nall et al. documented the expected decreas-es in the FA of cortical grey matter with the increasing postmenstrual age of the preterm neonates and then sought to establish which clinical parameters inf uenced cortical devel-opment af er preterm delivery. Although the authors’ term-equivalent cortical data could be compared with data from term-born con-trol neonates, a complete understanding of the normality of the observed decreases in cortical FA over time af er preterm delivery would ideally require a parallel study of nor-mal fetal cortical development in utero. Fetal dif usion studies have been performed, but in order to produce comparable data on cor-tical anisotropy, the imaging techniques re-quire considerable optimization to overcome the ef ects of maternal and fetal motion and of the inherently poor signal-to-noise ratio of the images.
MECHANISMS BEHIND THE MAYHEMAlthough it is accepted that cortical devel-opment is altered by preterm delivery, the causal mechanisms remain unclear. T ere are several alternative hypotheses. Accord-ing to the f rst hypothesis, cortical volume loss is a product of neuronal loss of function and death (primary neuronal degeneration), possibly accompanied by so-called second-ary retrograde neuro-axonal degeneration, which results from primary injury to imma-ture white matter and subplate, a common occurrence in preterm infants. T e second hypothesis is that cortical volume loss arises because of a failure of neuronal maturation rather than cell death or axonal degenera-tion. A third hypothesis—which implies that neither neuronal degeneration nor aborted maturation accounts for the brain volume decrease—suggests that thalamic areas of the brain (Fig. 1) are the primary sites that show altered growth in preterm infants (Fig. 2).
N E U R O L O G Y
Brain Maturation After Preterm BirthZoltán Molnár1* and Mary Rutherford2, 3
*Corresponding author. E-mail: [email protected]
1Department of Physiology, Anatomy and Genetics, University of Oxford, Le Gros Clark Building, South Parks Road, Oxford OX1 3QX, UK. 2Centre for the Developing Brain, Perinatal Imaging & Health, Imaging Sciences & Biomedical Engineering Division, King’s College Lon-don, London SE1 7EH, UK. 3Perinatal Imaging Group, Robert Steiner MR Unit, Medical Research Council (MRC) Clinical Sciences Centre, Imperial College, Ham-mersmith Hospital, London W12 OHS, UK.
P E R S P E C T I V E
Two translational studies—one in humans and one in sheep—suggest that (i) premature birth is associated with delayed maturation of grey matter in the cerebral cortex and (ii) medical care that prohibits impairment of growth in premature neonates may enhance cortical development and reduce neurological disabilities associated with preterm birth.
on
Janu
ary
16, 2
013
stm
.sci
ence
mag
.org
Dow
nloa
ded
from
P E R S P E C T I V E
www.ScienceTranslationalMedicine.org 16 January 2013 Vol 5 Issue 168 168ps2 2
T e thalamic region—a conglomerate of sev-eral nuclei with cortical connections that sit between the telencephalon and midbrain—relays information between the cerebral cor-tex and the sensory organs and subcortical regions of the brain. Altered growth in the thalami would result in secondary alterations in cortical development.
Premature newborns who display intra-uterine growth restriction (birth weights in the <10th percentile) demonstrate altered cortical folding and reduced cortical volumes when compared to preterm infants born at an appropriate weight for gestational age or full-term controls (5–7). T e studies behind these f ndings were signif cant at the time, but they all had shortcomings. Tof et al. (6) used sub-optimal criteria to def ne intrauterine growth restriction and did not acquire true volumet-ric MR sequences. T e Dubois et al. study (5) was conducted on a relatively small cohort (N = 35) that was further subdivided into infants
with (N = 10) and without brain lesions. T e authors assessed the folding process of the developing cortex rather than specif c brain volumetric parameters and showed an ad-ditional nonsignif cant alteration in cortical development: the presence of small white matter lesions. T e Tolsa et al. study (7) as-sessed the impact of intrauterine growth re-striction on cortical development in infants born preterm. T e authors used an adequate def nition of intrauterine growth restriction and a true volumetric MR sequence to quan-tify brain parameters but had no term-born control group: T e clinical neurological out-come was assessed only at term-equivalent age. T erefore, no relationships could be es-tablished between their brain parameters and later cognitive or behavioral scores.
Both postnatal somatic (body) growth and cortical development in neonates born very preterm (less than 30 weeks of gestation) have been directly associated with cognitive abil-
ity in later years (8, 9). Cerebral palsy (CP)—nonprogressive movement disorders caused by ischemic brain injury during pregnancy or birth—results from lesions in the brain, and it is unusual to f nd CP in children who were born preterm (so-called ex preterm) but do not show evidence of brain lesions or abnor-mality with high-quality MRI. However, ex preterm children may exhibit altered cortical connectivity and synchronization (assessed by functional MRI) during cognitive tasks rel-ative to full-term control children, even in the absence of signif cant cognitive disability (10, 11). T e etiology of altered cortical develop-ment and processing in ex preterm children is currently not known.
SIZE MATTERST e Vinall et al. (2) human MRI study set out to examine the extent to which poor postna-tal growth relates to the microstructure of the developing cerebral cortex in human infants
Fig. 1. Integration of newborn neurons. (A) Outline of a coronal section through an embryonic human brain. Shown at bottom is the left hemi-sphere with cortex, hippocampus, and thalamus indicated. The boxed area of the cortex is presented with more detail in (B). (B) Sectors of coronal sec-tions through the developing human cerebral cortex at two stages to illustrate the major anatomical changes in cortical organization during human brain development. Left, ~16 weeks postconception (16 PCW), peak of neurogenesis and cell migration; right, ~30 PCW, later when the cortical plate has matured further and neurogenesis is more advanced. Top, pial surface; bottom, ventricular surface. At the earlier stage (left), the germinal zone comprises several layers (ventricular zone and subventricular zone). These compartments contain the radial glia progenitor cells, which will generate the neurons of the cerebral cortex. The upper layers include the subplate and intermediate zones, the cortical plate, and the marginal zone. At this immature stage, there is a striking radial orientation of progenitor cells, migrating neurons, and newly born neurons. Most cortical plate neurons have a radially oriented apical dendrite reaching to the pial surface. At the later stage (right), the most signifi cant changes include the relative reduction of radial progenitors as neurogenesis is close to completion; elaboration of the somatodendritic morphology of the cortical and subplate neurons; and increasing aff erent and eff erent connections in the subplate and white matter and within the cortical plate. (Labels of progenitor cells and migrating neurons shown at left also apply to those on the right.) Note the extensive fi ber layer within subplate and intermediate zone and the periventricular fi ber layer within the subventricular zone. Most of the changes between the two stages increase the vertically oriented cellular processes. The small insets from the cortical plate illustrate the changes in somatodendritic morphology of the pyramidal neurons.
Marginal zone
Cortical plate
Ventricular
zone
Subventricular
zone
Subplate
Intermediate
zone
Cortical plate
neurons
Cortical plate
neurons
Subventricular
zone progenitor
Radial glia
progenitor cell
Outer radial glia
Subventricular zone
progenitor
Migrating
neuron
16 PCW 30 PCWA B
Cortex
Thalamus
Hippocampus
Pial surface
CR
ED
IT: Z
. MO
LNA
R; Y
. HA
MM
ON
D/S
CIE
NC
E T
RA
NSL
ATI
ON
AL
ME
DIC
INE
on
Janu
ary
16, 2
013
stm
.sci
ence
mag
.org
Dow
nloa
ded
from
P E R S P E C T I V E
www.ScienceTranslationalMedicine.org 16 January 2013 Vol 5 Issue 168 168ps2 3
born very preterm. In evaluating the relation-ship between neonatal (postbirth) growth and cortical development, it is important to consider the ef ects of multiple medical con-ditions in infants born very preterm that can confound experimental measurements in this cohort. T erefore, the authors included sev-eral such variables—evidence of intrauterine growth restriction, gestational age, and sex—in longitudinal multivariable models to relate postnatal growth restriction with cortical dif-fusion tensor imaging parameters.
T e study revealed that neonatal growth predicted cortical grey matter maturation independent of gestational age, birth weight percentile, brain injury, and systemic illness. Changes in cortical FA of infants born very preterm with impaired postnatal growth re-f ected changes in the radial, but not the axial, dif usion axis, suggesting a delay in the for-
mation of neuronal processes within the ce-rebral cortices. In contrast to the grey matter, FA values for the white matter were not as-sociated with postnatal weight changes, sug-gesting that white matter maturation is spared from ef ects of postnatal growth restriction.
Previous studies have emphasized that altered cortical development is associated with intrauterine growth restriction as a subsequent result of def cits in myelination. Although this association may still be valid, the Vinall et al. (2) study turns the focus from cerebral white matter development to neonatal somatic growth and its associ-ated cortical grey matter vulnerability. T e new work also implies that somatic growth restrictions that are more pronounced than expected for a particular gestational age and that persist af er birth are associated with lower than normal cerebral cortical vol-
umes and altered brain microstructure early in life. Extending the work of Vinall et al. to prenatal stages could yield mechanistic information about dif erent kinds of brain developmental aberrations. Intrauterine growth restriction that results from placen-tal dysfunction may have dif erent ef ects on the developing brain than the presence of poor postnatal growth in preterm infants, which is likely to be multifactorial.
T e results of Vinall et al. have important implications for future basic, translational, and clinical research on brain development and function. Neonatal growth over and above birth weight, brain injury, and system-ic illness all predict the extent of cortical grey matter maturation in infants who are assessed in neonatal intensive care units. T erefore, by establishing the optimal somatic growth rate for a preterm infant and how this is best
Fig. 2. Thalamocortical connections. Advanced diff usion tensor imaging study at 3 Tesla of a cohort of preterm-born human infants imaged at term-equivalent age (N = 47). Diff usion tractography was used to compare thalamocortical connectivity between preterm infants and healthy term-born controls (N = 18). The regions in the cortex where connections to the thalamus were signifi cantly altered after preterm birth are shown in red and yel-low. The larger the diff erence, the higher the value (yellow = highest). Cortical connectivity was calculated on a voxelwise basis, and the test statistics are corrected for multiple comparison across the cortex (P < 0.001). [Reproduced with permission from (16)]
CR
ED
IT: R
EPR
OD
UC
ED
WIT
H P
ER
MIS
SIO
N F
RO
M B
ALL
ET
AL.
(16)
on
Janu
ary
16, 2
013
stm
.sci
ence
mag
.org
Dow
nloa
ded
from
P E R S P E C T I V E
www.ScienceTranslationalMedicine.org 16 January 2013 Vol 5 Issue 168 168ps2 4
achieved, clinicians have the opportunity to optimize environmental conditions to mimic those that permit cortical development to proceed normally in infants born very pre-term. However, it remains dif cult to dissoci-ate isolated poor somatic growth from poor growth that occurs secondary to ongoing or earlier illnesses. Guidelines for optimal neo-natal growth also must address studies that provide evidence that accelerated postnatal growth increases the risks of subsequent ad-verse metabolic, endocrine, and cardiovascu-lar outcomes (12).
T e Vinall et al. study demonstrates the value of measuring FA to assess cortical mat-uration in preterm newborns and empha-sizes that FA of cortical grey matter decreases nonlinearly with increasing postmenstrual age. Numerous entities can lead to changes in MRI-def ned cortical microstructure, and the relationship between MRI f ndings and their anatomical correlates of brain matura-tion is not clear from this study (Fig. 1). In-deed, researchers lack a basic understanding of the remodeling of these circuits and their imaging correlates.
COUNTING SHEEPIt would be of great interest to systemati-cally compare various cortical areas during normal fetal development and then examine the ef ects of prematurity with or without acquired brain injury. Such studies require combined imaging and detailed anatomical investigations. T e new research by Dean et al. (3) aims to do just that. T is study made use of a fetal sheep model of preterm birth and revealed that prenatal cerebral ischemia reduced cortical volume and disrupted MRI-def ned cortical microstructure through reduction in the development of dendritic arborization of cortical neurons rather than from a reduction in neuron numbers. T e immature (0.65 gestation) fetal sheep is a global cerebral ischemia model from revers-ible bilateral carotid occlusion. Similar to hu-mans, the sheep model has limited cerebral vascular autoregulation (that is, unable to maintain adequate and stable cerebral blood f ow in conditions of over- or underperfu-sion), and it allows the monitoring of fetal heart rate, blood pressure, blood gases, and cerebral blood f ow. T e relative timing and general pattern of normal human and fetal sheep white-matter maturation and myelina-tion are similar to those of the ischemic fetal sheep model used by Dean et al., which cor-responded to an ~24- to 28-week-old human premature infant.
Dean and colleagues (3) show that fetal sheep that had undergone a single 37-min bilateral carotid artery occlusion inf icted at 91 days of gestation time displayed no loss of cortical neurons several weeks later af er premature birth. However, these sheep did display reduced dendritic arborization 4 weeks af er birth and an associated post-mortem reduction in cortical MR FA values measured at 11.7 T in f xed cortical tissue, which suggest disrupted cortical matura-tion. T e authors ruled out the possibil-ity of signif cant neuronal loss by assessing cortical neuronal death with a terminal deoxynucleotidyl transferase-mediated de-oxyuridine triphosphate nick end labeling (TUNEL) assay.
T e Dean et al. study (3) makes an im-portant contribution toward an under-standing of why there is an apparent loss of cortical volume in human preterm infants when assessed with clinical MRI. Until now, there has been inadequate data to determine whether cortical volume loss results from neuronal loss and, if so, whether this loss is a primary ef ect or a downstream ef ect from primary white matter and oligoden-drocyte loss, which was previously observed in the preterm sheep brain af er hypoxia by the same authors in their previous studies. However, despite providing some much needed answers, the new work leaves us with more questions. T e experimental data of Dean et al. are conf ned to the cortical plate. It is unfortunate that data are not pre-sented on the subplate, a transient zone in the developing cortex that contains some of the earliest generated neurons and connec-tions below the cortical plate (Fig. 1). Across the gestational window, the subplate is es-sential for setting up cortical neuronal syn-aptic networks while also being the subject of involution (of subplate neurons) through apoptosis (13). Even in other model organ-isms, it is not yet clear whether cortical den-dritic arborization may be driven in part by the (regressing) subplate.
Dean et al. (3) convincingly show distur-bances in dendritic arborization and spine elaboration in their sheep model and sug-gest that pyramidal neurons—the primary excitation units of the mammalian cor-tex—in the cortical plate are highly suscep-tible to abnormal development. T e authors used Golgi staining to expose the complete somatodendritic (cell body and all its pro-cesses) morphology of selected neurons in the frontal cortex at the level of dif use white matter injury in sheep brains four
weeks af er the hypoxic insult and analyzed dendritic length, branching complexity, and other measures of neuronal maturation. T e reduction in dendritic arborizations in the hypoxic-insult model compared to con-trols was distributed across all cortical lay-ers. Neuronal complexity was analyzed at multiple time points af er the hypoxic event (at this time, normal dendritic arbor com-plexity is low). Four weeks later, in control brains, the arbor underwent rapid expan-sion consistent with the pronounced expan-sion in cortical volume. T e authors suggest that dendritic abnormalities observed in the hypoxic-insult model are related to disrupt-ed maturation in this four-week time win-dow rather than to dieback of a previously formed arbor.
T is analysis was performed on recon-structed cells from Golgi-stained prepara-tions. Because of the nature of this analysis, the authors had to select pyramidal cells (which was performed blind) from stained specimens. In such comparisons, it is very important that the same cell types from the same layer and area are sampled and com-pared. T ere are hundreds of dif erent cell types estimated in the cortex, making this a dif cult task. T e authors make the point that the “ef ect of ischemia on neuronal complexity was independent of cortical lo-cation.” T e resolution of this statement was restricted to the supra- and infragranular layers of the cerebral cortex and was not ex-amined in specif c layers. Because matura-tion of the cortical f elds follows a specif c sequence, the study should be extended to several dif erent cell types and cortical re-gions.
T e disrupted dendritic maturation is consistent with the FA abnormalities ob-served 4 weeks af er ischemic insult in the sheep (3) and clinically in preterm infants (2). However, it is not clear whether all neu-ronal types are af ected or whether a partic-ular class of pyramidal neurons is relevant to the study of cognitive disabilities seen so frequently in preterm survivors. If not all pyramidal neurons are af ected, then what proportion of cortex pyramidal neuron ab-normalities is suf cient to result in altered FA across the whole cortex? And are the al-terations in arborization alone suf cient to result in the signif cant alterations in mea-surable cortical volumes demonstrated in preterm infants? Moreover, cortical circuit development is not a linear process in which simple, radially oriented elements progres-sively develop more and more vertically
on
Janu
ary
16, 2
013
stm
.sci
ence
mag
.org
Dow
nloa
ded
from
P E R S P E C T I V E
www.ScienceTranslationalMedicine.org 16 January 2013 Vol 5 Issue 168 168ps2 5
extending dendrites. T ere are numerous regressive events during normal brain de-velopment during which dendritic arbors are remodeled (for example, pyramidal neurons lose apical dendrites, spiny stellate cells remodel their apical dendrites, subplate neurons retract dendrites) (14). Examina-tion of these specif c cell populations would be an interesting future direction.
FORM AND FUNCTIONT e new studies (2, 3) highlight the need for basic research to provide mechanis-tic information on cerebral cortical circuit formation and the causes of cortical aber-rations. T e connectivities of several brain cell types change over time and are modeled according to precise phases of cortical cir-cuit assembly, and many of these processes have not been deciphered even in the most common rodent model systems. We know even less about these changes when the de-velopmental program is altered by external factors and the subsequent developmental program is derailed. T e Dean et al. study (3) represents an example of the way for-ward: It employs a model that is, in many respects, similar to the human situation, but modern developmental biology and imag-ing tools can be used to gain mechanistic insights. T e authors establish the f rst links between f ne changes in cortical circuits (so-matodentritic remodeling, spine formation, and progressive and regressive events) and MRI data in the same model. T e current resolution obtainable with clinical imaging might not be suf cient to detect many of these changes, but it can begin to link clini-cal f ndings such as those of Vinall et al. (2) to mechanistic explorations in model sys-tems. Linking cortical MRI parameters with functional MRI studies in preterm infants could help researchers ascertain the cor-responding functional signif cance of these parameters for early and subsequent neuro-cognitive development.
T e two new studies also illustrate the need for more intricate imaging of animals and patients and how mechanistic studies in animals can shed light on the human con-dition. T e reduction in the development of dendritic arborization of cortical neurons characterized in the sheep model provides a new focus on the role of the environment in improving adverse neurodevelopmental outcomes in preterm newborn infants. It is possible that specif c stimulation of the neo-nate may improve the development of den-dritic arborization. Stimulus-driven altera-tions in brain development might explain the major observed ef ects of social class and maternal education on long-term cognitive outcomes (15).
REFERENCES AND NOTES 1. H. Stolp, A. Neuhaus, R. Sundramoorthi, Z. Molnár, The
long and the short of it: Gene and environment inter-actions during early cortical development and conse-quences for long-term neurological disease. Front. Psy-chiatry 3, 50 (2012).
2. J. Vinall, R. E. Grunau, R. Brant, V. Chau, K. J. Poskitt, A. R. Synnes, S. P. Miller, Slower postnatal growth is associated with delayed cerebral cortical maturation in preterm newborns. Sci. Transl. Med. 5, 168ra8 (2013).
3. J. M. Dean, E. McClendon, K. Hansen, A. Azimi-Zonooz, K. Chen, A. Riddle, X. Gong, E. Sharifnia, M. Hagen, T. Ah-mad, L. A. Leigland, A. R. Hohimer, C. D. Kroenke, S. A. Back, Prenatal cerebral ischemia disrupts MRI-defi ned cortical microstructure through disturbances in neuro-nal arborization. Sci. Transl. Med. 5, 168ra7 (2013).
4. G. Ball, J. P. Boardman, D. Rueckert, P. Aljabar, T. Arichi, N. Merchant, I. S. Gousias, A. D. Edwards, S. J. Counsell, The eff ect of preterm birth on thalamic and cortical develop-ment. Cereb. Cortex 22, 1016–1024 (2012).
5. J. Dubois, M. Benders, A. Cachia, F. Lazeyras, R. Ha-Vinh Leuchter, S. V. Sizonenko, C. Borradori-Tolsa, J. F. Mangin, P. S. Hüppi, Mapping the early cortical folding process in the preterm newborn brain. Cereb. Cortex 18, 1444–1454 (2008).
6. P. B. Toft, H. Leth, P. B. Ring, B. Peitersen, H. C. Lou, O. Hen-riksen, Volumetric analysis of the normal infant brain and in intrauterine growth retardation. Early Hum. Dev. 43, 15–29 (1995).
7. C. B. Tolsa, S. Zimine, S. K. Warfi eld, M. Freschi, A. Sancho Rossignol, F. Lazeyras, S. Hanquinet, M. Pfi zenmaier, P. S. Huppi, Early alteration of structural and functional brain development in premature infants born with intrauter-ine growth restriction. Pediatr. Res. 56, 132–138 (2004).
8. I. G. Streimish, R. A. Ehrenkranz, E. N. Allred, T. M. O’Shea, K. C. Kuban, N. Paneth, A. Leviton, ELGAN Study Inves-tigators, Birth weight- and fetal weight-growth restric-tion: Impact on neurodevelopment. Early Hum. Dev. 88, 765–771 (2012).
9. O. Kapellou, S. J. Counsell, N. Kennea, L. Dyet, N. Saeed, J. Stark, E. Maalouf, P. Duggan, M. Ajayi-Obe, J. Hajnal, J. M. Allsop, J. Boardman, M. A. Rutherford, F. Cowan, A. D. Edwards, Abnormal cortical development after prema-ture birth shown by altered allometric scaling of brain growth. PLoS Med. 3, e265 (2006).
10. Y. Gozzo, B. Vohr, C. Lacadie, M. Hampson, K. H. Katz, J. Maller-Kesselman, K. C. Schneider, B. S. Peterson, N. Ra-jeevan, R. W. Makuch, R. T. Constable, L. R. Ment, Altera-tions in neural connectivity in preterm children at school age. Neuroimage 48, 458–463 (2009).
11. G. A. Lodygensky, L. Vasung, S. V. Sizonenko, P. S. Hüppi, Neuroimaging of cortical development and brain con-nectivity in human newborns and animal models. J. Anat. 217, 418–428 (2010).
12. V. Vasu, N. Modi, Assessing the impact of preterm nutri-tion. Early Hum. Dev. 83, 813–818 (2007).
13. I. Kostovic, L. Vasung, Insights from in vitro fetal magnet-ic resonance imaging of cerebral development. Semin. Perinatol. 33, 220–233 (2009).
14. A. Hoerder-Suabedissen, Z. Molnár, Morphology of mouse subplate cells with identifi ed projection targets changes with age. J. Comp. Neurol. 520, 174–185 (2012).
15. H. S. Wong, P. Edwards, Nature or nurture: A systematic review of the eff ect of socio-economic status on the de-velopmental and cognitive outcomes of children born preterm. Matern. Child Health J. 8, 10.1007/s10995-012-1183-8 (2012).
16. G. Ball, J. P. Boardman, P. Aljabar, A. Pandit, T. Arichi, N. Merchant, D. Rueckert, A. D. Edwards, S. J. Counsell, The infl uence of preterm birth on the developing thalamocortical connectome. Cortex 10, 10.1016/j.cor-tex.2012.07.006 (2012).
17. N. Padilla, C. Falcón, M. Sanz-Cortés, F. Figueras, N. Bar-gallo, F. Crispi, E. Eixarch, A. Arranz, F. Botet, E. Gratacós, Diff erential eff ects of intrauterine growth restriction on brain structure and development in preterm infants: a magnetic resonance imaging study. Brain Res. 1382, 98–108 (2011).
18. R. A. McKinney, M. Capogna, R. Dürr, B. H. Gähwiler, S. M. Thompson, Miniature synaptic events maintain den-dritic spines via AMPA receptor activation. Nat. Neurosci. 2, 44–49 (1999).
Funding: Medical Research Council, The Wellcome Trust, BBSRC UK and EU FP7. Competing interests: The authors de-clare that they have no competing interests.
Citation: Z. Molnár, M. Rutherford, Brain maturation after pre-term birth. Sci. Transl. Med. 5, 168ps2 (2013).
10.1126/scitranslmed.3005379
on
Janu
ary
16, 2
013
stm
.sci
ence
mag
.org
Dow
nloa
ded
from
