Skip to main content
PMC home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2018 Aug 1.
Published in final edited form as: J Pediatr. 2017 Jun 2;187:58–65.e1. doi: 10.1016/j.jpeds.2017.04.059

Associations of newborn brain MRI with long-term neurodevelopmental impairments in very preterm children

Peter J Anderson 1,2, Karli Treyvaud 2,4, Jeffrey J Neil 5,6, Jeanie LY Cheong 2,7,8, Rodney W Hunt 2,3,9, Deanne K Thompson 2,3, Katherine J Lee 2,3, Lex W Doyle 2,3,7,8, Terrie E Inder 6,10
PMCID: PMC5533625  NIHMSID: NIHMS871923  PMID: 28583705

Abstract

Objective

To determine the relationship between brain abnormalities on newborn MRI and neurodevelopmental impairment at 7 years of age in very preterm children.

Study design

223 VP infants (<30 weeks’ gestation or <1250 g) born at Melbourne’s Royal Women’s Hospital had a brain MRI scan at term equivalent age. Scans were scored using a standardized system that assessed structural abnormality of cerebral white matter (CWM), cortical gray matter (CGM), deep gray matter (DGM), and cerebellum (CBL). Children were assessed at 7 years on measures of general intelligence, motor functioning, academic achievement, and behavior.

Results

186 VP children (83%) had both an MRI at term equivalent age and a 7-year follow-up assessment. Higher global brain, CWM, and DGM abnormality scores were related to poorer IQ (p’s < .01), spelling (p’s < .05), math computation (p’s < .01), and motor function (p’s < .001). Higher CBL abnormality scores were related to poorer IQ (p = .001), math computation (p = .018) and motor outcomes (p = .001). Perinatal, neonatal and social confounders had little effect on the relationships between the MRI abnormality scores and outcomes. Moderate-severe global abnormality on newborn MRI was associated with a reduction in IQ (−6.9 points), math computation (−7.1 points), and motor (−1.9 points) scores independent of the other potential confounders.

Conclusions

Structured evaluation of brain MRI at term equivalent is predictive of outcome at 7-years of age, independent of clinical and social factors.

Keywords: brain pathology, prematurity, outcome, brain imaging

Approximately 10% of very preterm children (VP; <32 weeks’ gestation) develop significant impairments, such as cerebral palsy, while an additional 50% develop cognitive, motor, academic or behavioral problems.(1) Perinatal factors related to adverse outcomes in VP children include lower gestational age (GA),(2) bronchopulmonary dysplasia,(3) infection,(4) intrauterine growth restriction,(5) moderate to severe brain injury on cranial ultrasound, (6) postnatal corticosteroid use,(7) and surgery as a newborn.(8) Despite the association of these clinical factors with adverse outcome, it remains challenging to predict impairment in individuals. A better understanding of the underlying nature of cerebral injury and altered brain development in VP infants may assist in identification of neurodevelopmental risk.

Magnetic resonance imaging (MRI) in the newborn period has improved awareness of brain injury and aberrant brain growth in VP infants(9–11) and may enhance the ability to predict neurodevelopmental outcomes. Common features on structural MRI in VP infants at term equivalent age include loss of white matter with enlarged lateral ventricles, signal abnormality in the white matter, delayed myelination, thinning of the corpus callosum, delayed cortical folding, and larger extracerebral space.(12, 13) We have reported that quantitative scoring of these abnormalities was related to cognitive and motor delay at 24 months, even after taking into account other medical risk factors.(14) Others have reported the associations of white matter abnormalities to cognitive, language, and motor deficits in older children using similar scoring systems.(15, 16) For a more comprehensive evaluation of the nature of brain abnormalities in the VP infant, we developed a new scoring system(17) that included an evaluation of the cerebellum and deep gray matter, which are both vulnerable to injury following VP birth. In addition, brain growth and ventricular size are measured, rather than subjectively assessed.

The current study aimed to determine the relationship between this more expanded objective scoring of structural brain abnormalities and neurodevelopmental outcome at 7 years of age in VP children. We hypothesized that newborn MRI abnormalities, including those relating to the deep gray matter (DGM) and cerebellum (CBL), would be associated with adverse school-aged outcomes independent of the effect of other prognostic perinatal variables.

METHODS

Participants comprised children born <30 weeks’ gestational age (GA) or <1250 g birth weight between July 2001 and December 2003 from the Royal Women’s Hospital in Melbourne, Australia. Two hundred and twenty-seven VP infants without a congenital abnormality known to affect development were originally recruited (67% recruitment rate); two infants later died, and two were later excluded due to a subsequent diagnosis of a congenital abnormality, leaving 223 VP infants.

All infants had a brain MRI as close as possible to their expected due date; those who had their scan between 38 and 42 weeks’ postmenstrual age were included in this study (n=211), of whom 186 (88%) were reviewed at age 7 years (9 withdrew, 10 declined 7-year assessment, 3 could not be contacted, and 3 had emigrated).

The study was approved by the Human Research Ethics Committees of the Royal Women’s Hospital and the Royal Children’s Hospital in Melbourne, Australia. Written informed consent was obtained from parents prior to data collection. Information on this follow-up study have been published previously.(18–21)

MRI

Infants were scanned without sedation in a 1.5-T General Electric MRI scanner (Signa LX Echospeed System; General Electric, Fairfield, Connecticut). Infants underwent T1-weighted (0.8- to 1.6-mm coronal slices; flip angle 45°; repetition time 35 ms; echo time 9 ms; field of view 21 × 15 cm2; matrix 256 × 192), and T2/proton density-weighted (1.7- to 3-mm coronal slices with axial and sagittal reconstructions at 3mm slices; repetition time 4000 ms; echo time 60/160 ms; field of view 22 × 16 cm2; matrix 256 × 192, interpolated to 512 × 512) sequences.

A standardized scoring system was used to assess the presence and severity of abnormalities in cerebral white matter (CWM), cortical gray matter (CGM), DGM, and CBL.(17) The system extends that described by Inder et al(12) by adding scales for assessing DGM and CBL, and integrates quantitative biometrics. The CWM scale (range 0–17) is the sum of six subscales assessing the presence and severity of cystic lesions, signal abnormality, myelination delay, thinning of the corpus callosum, lateral ventricle dilatation, and volume reduction. Scores <3 were categorized as normal, 3–4 were categorized as mild abnormality, and >4 were categorized as moderate to severe abnormality. The CGM scale (range 0–9) is the sum of three subscales assessing signal abnormality, delayed gyral maturation, and increased extracerebral space. The DGM and CBL scales (range from 0–7) have two subscales assessing signal abnormality and volume reduction. For the CGM, DGM, and CBL abnormality scales, scores of 0 were categorized as normal, 1 as mild abnormality, and >1 as moderate to severe abnormality. A global brain abnormality score (range 0–40) is generated by summing the CWM, CGM, DGM, and CBL scales. For the global brain scale, scores <4 were categorized as normal, 4–7 as mild abnormality, and >7 as moderate to severe abnormality. Scans were reviewed by an experienced neonatal neurologist independent of knowledge of long-term outcomes, with excellent inter-rater and intra-rater reliabilities (ICC > 0.90).(17)

Neurodevelopmental Assessment

At 7 years’ corrected age, general intelligence, academic achievement, motor functioning, and behavior were assessed. General intellectual functioning was assessed using the Full Scale IQ (FSIQ) of the Wechsler Abbreviated Scale of Intelligence (WASI).(22) Word reading, spelling, and math computation were assessed with the Wide Range Achievement Test (WRAT-4).(23) Motor skills were assessed using the Movement Assessment Battery for Children (MABC2).(24) Parents rated their child’s behavior using the Strengths and Difficulties Questionnaire (SDQ).(25) The total difficulty score of the SDQ was used as an estimate of behavioral problems, with a higher score reflecting greater behavioral difficulty (range: 0–40). Age standardized scores are reported for the WASI (Mean=100, SD=15), WRAT4 (Mean=100, SD=15) and MABC2 (Mean=10, SD=3) based on the child’s corrected age to avoid bias in cognitive test scores.(26) The SDQ does not provide age standardized scores, and raw scores are reported. Children who did not complete the test because it was too difficult were assigned a score that was 3 SD below the normative mean for that test (or above 1 SD for the SDQ). Assessments were performed by trained assessors who had no knowledge of the child’s medical history or newborn MRI.

Outcome Risk Factors

Perinatal data were obtained from chart review and socio-demographic information was obtained from a caregiver questionnaire. Birth factors include antenatal corticosteroid exposure, multiple birth, sex, GA, and birth weight standardized for GA and sex (birth weight Z-score).(27) Neonatal factors included grade 3 or 4 intraventricular hemorrhage (IVH), cystic periventricular leukomalacia (PVL), bronchopulmonary dysplasia (BPD; defined as the requirement for oxygen at 36 weeks’ postmenstrual age), postnatal corticosteroid exposure, infection (either proven sepsis or necrotizing enterocolitis (NEC)(28), and surgery in the newborn period. Social factors were assessed at age 7 years using a composite measure(29) assessing family structure, education of the primary caregiver, occupation and employment status of the primary income earner, language spoken at home, and maternal age when the child was born. Scores were categorized into lower (≤2) and higher (>2) social risk based on the median score of 2.

Statistical Analyses

Data were analyzed using Stata 14.(30) The relationships between newborn MRI scores and 7 year outcomes were examined by linear regression fitted using Generalized Estimating Equations (GEEs) with an exchangeable correlation structure and robust standard errors to allow for correlations between twins/triplets in the study.(31, 32) To investigate the effects of MRI scores independent of other perinatal, neonatal and social predictors, potential confounders of child outcomes were added to the regression models.

The association between MRI abnormality categories (normal, mild, moderate-severe) and neurodevelopmental outcome were also examined using linear regression fitted using GEEs with and without adjustment for other perinatal, neonatal and social predictors. Finally, the independent contribution of moderate-severe abnormality for the global scale was assessed with the other perinatal, neonatal and social predictors, again using multivariable regression.

There was no adjustment for multiple testing, but we acknowledge the many comparisons presented and have interpreted our findings by focusing on overall patterns and magnitude of differences, rather than on individual p-values.(33)

RESULTS

The neonatal characteristics of the 186 children are presented in Table 1. The characteristics of the 186 infants included in the study did not differ from the 25 not followed up at 7 years (Table 2; Online only). On newborn MRI, 68% of infants had at least a mild abnormality on the global score, while for the other scales the rate of any abnormality ranged from 40% (CGM) to 64% (CBL) (Table 1).

Table 1.

Characteristics of the VP cohort (n=186)

Variable Summary
Antenatal corticosteroids, % (n) 87 (164)
Multiple birth, % (n) 45 (83)
Gestational age (weeks), mean (SD) 27.5 (1.9)
Birth weight (g), mean (SD) 973 (220)
Birth weight Z-score, mean (SD) −0.54 (0.94)
Male, % (n) 52 (96)
Grade 3–4 intraventricular hemorrhage, % (n) 4 (7)
Cystic periventricular leukomalacia, % (n) 4 (7)
Bronchopulmonary dysplasia, % (n) a 33 (61)
Postnatal corticosteroids, % (n) 8 (14)
Infection (proven sepsis or necrotizing enterocolitis), % (n) 35 (66)
Surgery in newborn period, % (n) 18 (34)
Gestational age (weeks) at MRI, mean (SD) 40 (1.05)
Social risk at age 7 years, median (IQR) 2 (1,3)
Newborn MRI scales
Global, median (IQR) (possible range 0–40) 5 (3,8)
 Normal (<4), % (n) 32 (60)
 Mild abnormality (4–7), % (n) 42 (78)
 Moderate/severe abnormality (>7), % (n) 26 (48)
Cerebral white matter, median (IQR) (possible range 0–17) 3 (1, 4)
 Normal (<2), % (n) 45 (83)
 Mild abnormality (3–4), % (n) 36 (67)
 Moderate/severe abnormality (>5), % (n) 19 (36)
Cortical gray matter, median (IQR) (possible range 0–9) 0 (0,1)
 Normal (0), % (n) 60 (111)
 Mild abnormality (1), % (n) 22 (41)
 Moderate/severe abnormality (>1), % (n) 18 (34)
Deep gray matter, median (IQR) (possible range 0–7) 1 (0,1)
 Normal (0), % (n) 44 (81)
 Mild abnormality (1), % (n) 32 (60)
 Moderate/severe abnormality (>1), % (n) 24 (45)
Cerebellar, median (IQR) (possible range 0–7) 1 (0,2)
 Normal (0), % (n) 36 (67)
 Mild abnormality (1), % (n) 38 (70)
 Moderate/severe abnormality (>1), % (n) 26 (49)
7-year Neurodevelopmental outcomes
Full scale IQ (n = 186), Mean (SD) 96 (15)
Word Reading (n = 182), Mean (SD) 98 (20)
Spelling (n = 182), Mean (SD) 98 (20)
Math computation (n = 182), Mean (SD) 89 (18)
Motor (n = 182), Mean (SD) 7.9 (3.8)
Behavior (n = 170), Mean (SD) 10.5 (6.4)
Open in a new tab
a

Missing in 1 child);

SD = standard deviation; IQR = inter-quartile range (25th – 75th centiles. Percentages of those with available data.

Table 2.

Online only. Neonatal characteristics of participants and non-participants.

Variable Participants
(n = 186)		 Non-participants
(n = 25)		 p
Multiple birth, % (n) 45 (83) 36 (9) .41
Gestational age (weeks), mean (SD) 27.5 (1.9) 27.8 (1.8) .49
Birth weight (g), mean (SD) 972.7 (219.9) 973.9 (258.1) .97
Birth weight Z–score, mean (SD) −0.5 (0.9) −0.7 (1.0) .49
Male, % (n) 52 (96) 36 (9) .14
Grade 3–4 intraventricular hemorrhage, % (n) 4 (7) 4 (1) .95
Cystic periventricular leukomalcacia, % (n) 4 (7) 4 (1) .95
Bronchopulmonary dysplasia, % (n) a 33 (61) 20 (5) .20
Postnatal corticosteroids, % (n) a 8 (14) 12 (3) .45
Infection (proven sepsis/necrotizing enterocolitis), % (n) 35 (66) 44 (11) .41
Gestational age (weeks) at MRI, mean (SD) 40 (1.05) 40 (1.15) .79
Global MRI score, median (IQR) 5 (3, 8) 5 (4, 9) .62
Cerebral white matter MRI score, median (IQR) 3 (1, 4) 3 (2, 4.5) .23
Cortical gray matter MRI score, median (IQR) 0 (0, 1) 0 (0, 1) .72
Deep gray matter MRI score, median (IQR) 1 (0, 1) 1 (0, 2) .77
Cerebellar MRI score, median (IQR) 1 (0, 2) 1 (0, 1) .34
Open in a new tab
a

Missing in 1 child;

SD = standard deviation; IQR = inter-quartile range (25th – 75th centiles); percentages of those with available data; p values reflect results from chi2/t-tests as appropriate.

On univariable analysis, higher abnormality scores for all MR variables were related to poorer IQ, reading, spelling, math computation, and motor function, but the evidence was stronger for some relationships and only weak for others (Table 3, unadjusted). There was little evidence that MRI abnormality scores were related to behavior. On multivariable analysis (Table 3, adjusted results), perinatal, neonatal and social confounders had little effect on the relationships between the MRI abnormality scores and outcomes; the estimated regression coefficients and 95% CIs were minimally affected and the evidence remained strong for most associations that were present on the univariable analyses (Table 3).

Table 3.

Unadjusted and adjusted linear associations between newborn MRI scales and neurodevelopmental outcomes at age 7 years.

Predictor IQ (n=186) Word Reading (n=182) Spelling (n=182) Math Computation (n=182) Motor (n=182) Behavior (n=170)
Est (95% CI) p Est (95% CI) P Est (95% CI) p Est (95% CI) p Est (95% CI) p Est (95% CI) p
Global score
Unadjusted −1.6 (−2.3, −0.9) <.001 −1.1 (−2.1, −0.1) .026 −1.4 (−2.4, −0.5) .003 −1.8 (−2.6, −1.0) <.001 −0.4 (−0.6, −0.3) <.001 0.3 (−0.0, 0.6) .096
Adjusteda −1.4 (−2.1, −0.7) <.001 −0.8 (−1.8, 0.3) .162 −1.1 (−2.1, −0.1) .029 −1.6 (−2.4, −0.8) <.001 −0.4 (−0.5, −0.2) <.001 0.2 (−0.1, 0.5) .281
White matter
Unadjusted −1.9 (−3.3, −0.5) .007 −1.6 (−3.4, 0.1) .072 −1.9 (−3.6, 0.17) .031 −2.2 (−3.6, −0.8) .002 −0.5 (−0.8, −0.3) <.001 0.3 (−0.1, 0.8) .175
Adjusteda −1.7 (−3.1, −0.4) .013 −1.4 (−3.1, 0.4) .119 −1.5 (−3.2, 0.2) .080 −2.0 (−3.4, −0.6) .006 −0.5 (−0.7, −0.2) .001 0.2 (−0.2, 0.7) .262
Cortical gray matter
Unadjusted −0.91 (−2.8, 1.0) .347 −0.02 (−2.2, 2.2) .980 −0.30 (−3.0, 2.4) .825 −0.5 (−3.1, 2.1) .698 −0.5 (−0.9, −0.04) .034 0.1 (−0.6, 0.9) .716
Adjusteda −0.0 (−2.0, 2.0) .976 −0.0 (−2.1, 2.1) .986 0.1 (−2.3, 2.5) .962 0.1 (−2.6, 2.8) .959 −0.2 (−0.6, 0.3) .409 0.0 (−0.7, 0.8) .916
Deep gray matter
Unadjusted −5.4 (−7.7, −3.1) <.001 −4.7 (−7.9, −1.4) .005 −5.5 (−8.8, −2.3) .001 −5.7 (−8.3, −3.0) <.001 −1.2 (−1.7, −0.6) <.001 0.7 (−0.3, 1.7) .178
Adjusteda −4.8 (−7.1, −2.5) <.001 −3.6 (−7.2, −0.1) .045 −5.2 (−8.7, −1.7) .003 −5.5 (−8.3, −2.8) <.001 −1.0 (−1.5, −0.4) <.001 0.2 (−0.7, 1.1) .676
Cerebellar
Unadjusted −3.6 (−5.7, −1.5) .001 −1.0 (−3.8, 1.8) .503 −2.3 (−5.0, 0.3) .087 −3.3 (−6.1, −0.6) .018 −1.0 (−1.5, −0.4) .001 0.9 (−0.1, 1.8) .088
Adjusteda −1.7 (−4.1, 0.6) .142 2.3 (−0.8, 5.4) .147 0.6 (−2.3, 3.4) .698 −0.6 (−3.5, 2.2) .661 −0.7 (−1.2, −0.1) .023 0.5 (−0.4, 1.5) .287
Open in a new tab

Est = estimate of regression coefficient (unit change in outcome variable per unit change in MR score) from separate linear regression models fitted using generalized estimating equations to allow for clustering of twins. CI = confidence interval;

a

Adjusted for antenatal corticosteroids, multiple birth, sex, gestational age, birth weight Z-score, grade 3–4 intraventricular hemorrhage, cystic periventricular leukomalacia, bronchopulmonary dysplasia postnatal corticosteroids, infection, surgery, and higher social risk.

In general, the outcomes were worse with increasing severity of brain abnormality (Table 4). On the global brain abnormality scale, neurodevelopmental outcome did not differ greatly between children in the normal and mild abnormality groups, but those with moderate-severe abnormality exhibited substantially poorer IQ, math, and motor skills. For the DGM abnormality scale, the differences between the normal and mild abnormality groups were wider on academic measures and motor functioning than for other abnormality scales, but the moderate-severe abnormality group had the poorest outcomes. The major finding on the CBL abnormality scale was for motor outcome, with poorer performance observed in the moderate-severe abnormality group. The abnormality groups on the CGM scale did not differ on long-term neurodevelopment measures. After adjusted for confounding variables, strong evidence persisted for i) poorer performance in IQ, math computation, and motor function with worse global abnormality scores, ii) poorer motor performance with worse CWM abnormality scores, iii) poorer IQ, spelling, math, and motor function with worse DGM abnormality scores, and iv) poorer motor performance with worse CBL abnormality scores.

Table 4.

Means and standard deviations for each child outcome by category of brain abnormality (normal, mild, moderate-severe)

MR score Normal
M (SD)		 Mild
M (SD)		 Mod-Severe
M (SD)		 p-valuea Adjusted p-valueb
Global n= 60 n= 78 n= 48
IQ 100.2 (14.7) 96.6 (14.8) 88.9 (15.2) <.001 .009
Word Reading 100.7 (19.1) 99.6 (19.1) 93.8 (20.8) .043 .522
Spelling 100.9 (18.9) 97.8 (18.9) 93.5 (21.1) .025 .323
Math Comp 92.9 (18.8) 90.5 (17.4) 81.7 (17.0) .001 .039
Motor 9.5 (3.7) 7.8 (3.6) 6.1 (3.3) <.001 .001
Behavior 10.2 (7.8) 10.0 (5.4) 11.8 (5.5) .306 .805
White Matter n= 83 n= 67 n= 36
IQ 98.4 (16.3) 95.9 (12.5) 89.6 (16.8) .007 .051
Word Reading 100.2 (20.3) 98.6 (17.7) 94.2 (21.3) .109 .328
Spelling 99.3 (20.5) 97.9 (16.6) 93.8 (22.5) .138 .319
Math Comp 91.5 (18.7) 89.2 (16.9) 83.1 (18.5) .020 .070
Motor 8.7(4.0) 8.2 (3.3) 5.6 (3.3) <.001 .003
Behavior 10.4 (7.2) 10.1 (5.8) 11.6 (5.3) .441 .906
Cortical Gray n= 111 n= 41 n=34
IQ 96.8 (14.4) 93.2 (14.7) 95.5 (19.1) .594 .645
Word Reading 99.7 (18.8) 94.7 (20.5) 98.7 (21.3) .820 .924
Spelling 98.7 (18.5) 95.2 (18.5) 97.7 (24.1) .996 .856
Math Comp 89.9 (17.4) 87.8 (17.5) 87.8 (21.8) .707 .942
Motor 8.4 (3.6) 7.5 (3.8) 6.9 (4.2) .059 .742
Behavior 10.3 (6.7) 10.0 (5.6) 11.7 (5.9) .567 .758
Deep Gray n=81 n=60 n=45
IQ 99.9 (14.3) 96.5 (14.5) 87.4 (15.5) <.001 <.001
Word Reading 103.3 (17.9) 95.2 (19.7) 93.7 (21.0) .004 .053
Spelling 103.7 (18.0) 94.5 (18.9) 91.0 (20.6) .001 .003
Math Comp 95.6 (18.4) 84.5 (16.5) 83.0 (16.3) <.001 <.001
Motor 9.1 (3.7) 7.5 (3.5) 6.4 (3.8) <.001 .002
Behavior 10.2 (7.0) 10.2 (5.6) 11.6 (6.0) .309 .797
Cerebellum n=67 n=70 n=49
IQ 99.2 (15.6) 96.3 (15.8) 90.4 (13.3) .001 .115
Word Reading 99.0 (19.4) 99.8 (19.1) 95.9 (20.8) .400 .297
Spelling 99.8 (19.0) 98.5 (20.1) 93.7 (19.5) .062 .995
Math Comp 91.7 (19.6) 89.0 (17.3) 85.4 (17.0) .010 .497
Motor 8.7 (3.8) 8.1 (3.8) 6.5 (3.4) .001 .029
Behavior 9.8 (6.5) 10.8 (6.5) 11.1 (6.0) .191 .350
Open in a new tab

Note.

a

p value from regression equation predicting outcome by brain abnormality group (none versus mild or moderate-severe);

b

Adjusted for antenatal corticosteroids, multiple birth, sex, gestational age, birth weight Z-score, grade 3–4 intraventricular hemorrhage, cystic periventricular leukomalacia, bronchopulmonary dysplasia, postnatal corticosteroids, infection, surgery, and higher social risk.

Moderate-severe global abnormality on newborn MRI was one of four perinatal and neonatal factors that were independently associated with neurodevelopment at 7 years, with the others being birthweight Z-score <−2 SD, postnatal corticosteroids, and newborn surgery (Table 5). Birth weight Z-score <−2 SD was independently associated with spelling and math, but this applied to only 16 children. Postnatal corticosteroids were independently associated with poorer IQ, spelling, and math, but were administered to only 14 children. Newborn surgery was associated with motor outcome only. Moderate-severe global abnormality on newborn MRI (48 children) was independently associated with an almost 0.5 SD reduction in IQ and math, and a greater than 0.6 SD reduction in motor outcome. In contrast, neither grade 3–4 IVH nor cystic PVL were independently associated with any 7-year outcomes. Higher social risk was associated only with lower IQ.

Table 5.

Independent contributions of perinatal, neonatal and social variables on neurodevelopmental outcomes at 7 years on multivariable analysis.

Neurodevelopmental outcomes
IQ (n=178) Word Reading (n=174) Spelling (n=174) Math Computation (n=174) Motor (n=175)
Predictor Est (95% CI) p Est (95% CI) P Est (95% CI) p Est (95% CI) P Est (95% CI) p
Moderate-severe global brain abnormality (n = 48) −6.9 (−11.8, −2.1) 0.005 −3.0 (−10.2, 4.1) 0.407 −1.4 (−8.0, 5.2) 0.678 −7.1 (−12.8, −1.3) 0.016 −1.9 (−3.1, −0.6) 0.003
Antenatal corticosteroids (n = 164) −0.01 (−5.4, 5.4) 0.995 −4.9 (−15.1, 5.4) 0.353 −3.5 (−14.2, 7.2) 0.519 −0.6 (−9.7, 8.5) 0.896 −0.1 (−1.7, 1.4) 0.858
Multiple (n = 83) −1.6 (−6.2, 3.0) 0.489 −0.7 (−7.1, 5.8) 0.842 −0.04 (−6.2, 6.1) 0.988 −0.6 (−6.4, 5.2) 0.841 0.1 (−1.1, 1.2) 0.921
GA <26 weeks (n = 33) 1.6 (−5.7, 8.9) 0.668 −1.9 (−11.8, 7.9) 0.698 −3.0 (−11.7, 5.6) 0.491 1.4 (−6.1, 9.0) 0.712 −1.1 (−2.8, 0.6) 0.194
Birth weight Z-score <−2 (n = 16) −5.8 (−13.5, 2.0) 0.143 −8.6 (−19.1, 1.8) 0.104 −10.5 (−20.0, −1.0) 0.030 −9.8 (−18.1, −1.5) 0.020 −1.6 (−3.7, 0.5) 0.135
Male (n = 96) −3.2 (−7.2, 0.9) 0.126 −2.1 (−7.8, 3.6) 0.469 −3.9 (−9.6, 1.7) 0.173 −0.2 (−5.4, 5.0) 0.940 −0.6 (−1.6, 0.5) 0.296
Grade 3–4 IVH (n = 7) −1.5 (−18.3, 15.3) 0.860 5.8 (−11.5, 23.1) 0.511 13.4 (−5.9, 32.7) 0.175 8.1 (−5.8, 22.0) 0.254 −1.5 (−4.1, 1.1) 0.273
Cystic PVL (n = 7) 1.1 (−9.7, 11.9) 0.846 −4.8 −19.9, 10.3) 0.533 −12.2 (−26.4, 2.0) 0.092 −6.8 (−16.2, 2.6) 0.156 0.7 (−1.4, 2.8) 0.525
BPD (n = 61) −3.5 (−9.7, 2.7) 0.269 −1.7 (−9.4, 5.9) 0.657 −2.6 (−9.7, 4.4) 0.461 −1.7 (−8.5, 5.0) 0.612 0.1 (−1.4, 1.6) 0.882
Post-natal corticosteroids (n = 14) −14.0 (−23.9, −4.1) 0.006 −11.4 (−25.8, 3.1) 0.122 −13.9 (−27.3, −0.4) 0.043 −12.6 (−22.4, −2.8) 0.011 −1.8 (−4.3, 0.7) 0.153
Infection (n = 66) −2.2 (−6.7, 2.2) 0.329 −1.7 (−7.9, 4.4) 0.577 −5.5 (−11.1, 0.2) 0.058 −3.1 (−8.4, 2.2) 0.253 0.4 (−0.8, 1.5) 0.517
Surgery prior to discharge (n = 34) −2.4 (−9.1, 4.3) 0.486 −3.7 (−13.6, 6.2) 0.469 −1.3 (−10.6, 8.0) 0.784 −4.1 (−12.4, 4.2) 0.328 −2.1 (−3.5, −0.6) 0.004
Higher social risk at age 7 years (n = 62) −4.7 (−9.2, −0.1) 0.043 −5.7 (−12.3, 0.9) 0.093 −4.6 (−10.7, 1.5) 0.140 −4.5 (−10.3, 1.3) 0.131 −0.6 (−1.9, 0.6) 0.312
Open in a new tab

Note. Est = estimate of regression coefficient from separate linear regression models fitted using generalized estimating equations to allow for clustering of twins; CI = confidence interval; BPD = bronchopulmonary dysplasia; IVH = intraventricular hemorrhage; PVL = periventricular leukomalacia. Bolded values indicate P<0.05

DISCUSSION

Our findings demonstrate that in children born very preterm abnormalities on newborn MRI are related to adverse neurodevelopmental outcomes at 7 years beyond the effect of other perinatal and neonatal variables. This highlights three key points: 1) brain abnormalities that develop by term equivalent gestational age have a lasting impact on cognitive, motor, and academic abilities; 2) MRI is useful for identifying abnormalities of importance in the preterm infant; and 3) a systematic, structural scoring system has advantages for the evaluation of structural MRI studies in this population. We previously reported that newborn MRI abnormalities, white matter abnormalities, predicted cognitive and motor development at age two years.(14) Although predicting early developmental delay is important, predicting longer-term outcomes is more challenging and important. Studies have demonstrated that white matter abnormalities on newborn MRI are related to IQ, language, executive function, and motor skills in pre-school and school-aged preterm children.(15, 16, 20, 34) However, these studies focused predominantly on the integrity of white matter and used a subjective MRI scoring system. The current study adds to these previous studies by utilizing a more comprehensive and objective scoring system.(17) Furthermore, our scoring system assesses the integrity of the D G M and CBL, which is relevant due to the sensitivity of these structures to injury following preterm birth,(11, 35) and their role in cognition, motor function, and behavior.(36–39) Neuronal loss and gliosis have been reported within the basal ganglia, thalamus and cerebellum in VP infants, especially in infants with white matter injury.(40) Deep gray abnormalities were identified in 56% of our sample, which were associated with lower IQ, reading, spelling, math, and motor outcomes, even after adjustment for perinatal, neonatal and social factors. Cerebellar abnormalities were also common (64%), and remained associated with poorer motor outcome, after considering potential confounders.

Preterm brain injury is a complex amalgam of cerebral injury and associated derangement with secondary trophic and maturational effects.(11) Multifaceted gray and white matter lesions reflect acquired insults, altered developmental trajectories, and reparative phenomena in various combinations. Given complex interactions between brain regions, alterations in the volume of a area may reflect primary injury to that area and/or be secondary to de-afferentation caused by white matter injury, particularly for regions with widespread connections such as the deep nuclear gray matter and cerebellum.(11) By virtue of these connections, alterations in CBL and DGM volume may essentially reflect overall injury to the white matter and cortex. Conversely, primary injury to the CBL may result in subsequent underdevelopment of the cerebral cortex due to reciprocal connections with cortex.(41) In the present study, DGM abnormality was particularly predictive of adverse neurodevelopmental outcome. The basal ganglia and thalamus are brain relay structures known to modulate a range of motor, cognitive, and sensory functions,(37, 39) and are vulnerable to preterm birth because of their marked developmental trajectory in the third trimester.(42) Previous studies have also reported smaller basal ganglia and thalamic volumes in VP children,(43–45) and an association between DGM volume and cognitive (43, 46) and motor (47, 48) functioning.

Surprisingly, CGM abnormality was not associated with the long-term outcomes examined, despite being significantly related to both cognitive and motor development at 24 months of age.(14) It could be that this pattern of abnormality, which is subjective in nature, may reflect a developmental lag rather than irreversible cortical injury. Alternatively, early CGM abnormality may be related to specific outcomes not examined in this paper, such as attention or executive function.(18)

A global brain MRI abnormality score, as used in this study, may provide the most accurate reflection of the overall neural consequences of preterm birth, and therefore, be more predictive of later neurodevelopmental impairment. Indeed, the global abnormality score was independently associated with all outcomes assessed except behavior. In contrast to infants with no or mild abnormality, infants with moderate to severe abnormality on this global scale (n=48; 26%) performed 0.5 to 0.6 SD lower on tests of IQ, math, and motor function after controlling for confounders. Although postnatal corticosteroids and birth weight Z-score <−2 SD were also independent predictors of neurodevelopmental outcome at 7-years, less than 10% of children had these risk factors. The global MRI scale was also more strongly associated with 7-year outcome than higher social risk. However, it is feasible that the influence of social risk may increase with age.(49) In summary, global abnormality on newborn MRI may help identify VP children who are most in need of early intervention.

Major IVH and PVL have been noted in the past to be strongly associated with adverse outcome.(50) The small number of subjects in our sample with IVH (n=7) and PVL (n=7), and hence low statistical power, may explain why these clinical factors were not independently related to 7-year outcomes. However, our findings do emphasize that more subtle abnormalities observed on newborn MRI are much more common than these severe brain ultrasound findings,(51) and suggest that newborn MRI provides important additional information to traditional prognostic variables.

This cohort was born between 2001 and 2003, and replication of this study in the current era is required, given ongoing advances in MRI technology and management of VP infants. We achieved excellent retention in this prospective, longitudinal study, and those lost to follow-up were similar in perinatal characteristics, increasing our confidence that we did not systematically fail to follow up higher-risk children. Another strength is that the MRI scoring system used can be applied in most settings with access to newborn MRI facilities and does not require advanced computer analysis. In contrast, advanced quantitative techniques such as cortical morphometry, shape analysis, tractography, and structural and functional connectivity are research applications, although they may become available to clinicians in the future. In this study, we focused on general functional domains such as IQ, academic, motor skills, and behavior, but further research is needed on more specific outcome domains. Although this is a large study of newborn MRI scanning with data on school-age outcomes, our sample size does limit power to identify subtle effects. Furthermore, we did not adjust for multiple comparisons as this was an exploratory study; instead we focused on patterns and strengths of associations rather than individual p-values.

This study confirms that newborn MRI identifies brain abnormalities in CWM, DGM, and CBL that have long-term impact on neurodevelopmental outcomes, independent of perinatal and social risk factors. Thus, quantitative evaluation of structural MRI obtained at term equivalent age provides valuable information for clinicians. Because discussion of neurodevelopmental prognosis with families prior to NICU discharge is standard of care, and brain abnormality on MRI is the strongest neonatal predictor of long-term outcome, prognostic discussions with families should be informed about MRI findings alongside other clinical indicators. For research, newborn MRI could be applied as a surrogate, shortterm outcome measure for neuroprotection studies while awaiting long-term follow-up data.

Acknowledgments

Supported by Australia’s National Health and Medical Research Council, including Centre for Clinical Research Excellence (546519), Centre for Research Excellence (1060733, 237117, and 491209), Senior Research Fellowship (1081288 [to P.A.]), Career Development Fellowship (1085754 [to D.T.]), Early Career Fellowship (1012236 [to D.K]), and Early Career Fellowship (1053787 [to J.C.]). Also supported by the Victorian Government’s Operational Infrastructure Support Program and the US National Institutes of Health (HD058056).

We thank Merilyn Bear, BN (Murdoch Childrens Research Institute), for recruitment; Michael Kean, RT, and the radiographers at Melbourne Children’s MRI Centre at the Murdoch Childrens Research Institute; and the Victorian Infant Brain Studies (VIBeS) and Developmental Imaging groups at the Murdoch Childrens Research Institute for their ideas and support, as well as the families and children who participated in this study.

Abbreviations

CBL

Cerebellum

CGM

cortical gray matter

CWM

cerebral white matter

DGM

deep gray matter

GA

gestational age

GEE

Generalising Estimating Equations

IQ

intelligence quotient

IVH

intraventricular hemorrhage

MABC2

Movement Assessment Battery for Children2nd Edition

MRI

magnetic resonance imaging

PVL

periventricular leukomalacia

SDQ

Strengths & Difficulties Questionnaire

VP

very preterm

WASI

Wechsler Abbreviated Scale of Intelligence

Footnotes

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

The authors have no conflicts of interest.

Edited by PF and WFB

References

  • 1.Hutchinson EA, De Luca CR, Doyle LW, Roberts G, Anderson PJ, for the Victorian Infant Collaborative Study Group (VICS) School-age outcomes of extremely preterm or extremely low birth weight children. Pediatrics. 2013;131:E1053–E61. doi: 10.1542/peds.2012-2311. [DOI] [PubMed] [Google Scholar]
  • 2.Wolke D, Strauss VYC, Johnson S, Gilmore C, Marlow N, Jaekel J. Universal gestational age effects on cognitive and basic mathematic processing: 2 cohorts in 2 countries. Journal of Pediatrics. 2015;166:1410–6.e1. doi: 10.1016/j.jpeds.2015.02.065. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3.Doyle LW, Anderson PJ. Long-term outcomes of bronchopulmonary dysplasia. Seminars in Fetal & Neonatal Medicine. 2009;14:391–5. doi: 10.1016/j.siny.2009.08.004. [DOI] [PubMed] [Google Scholar]
  • 4.van Vliet EOG, de Kieviet JF, Oosterlaan J, van Elburg RM. Perinatal infections and neurodevelopmental outcome in very preterm and very low-birth-weight infants: A meta-analysis. JAMA Pediatrics. 2013;167:662–8. doi: 10.1001/jamapediatrics.2013.1199. [DOI] [PubMed] [Google Scholar]
  • 5.Murray E, Fernandes M, Fazel M, Kennedy SH, Villar J, Stein A. Differential effect of intrauterine growth restriction on childhood neurodevelopment: a systematic review. BJOG-an International Journal of Obstetrics and Gynaecology. 2015;122:1062–72. doi: 10.1111/1471-0528.13435. [DOI] [PubMed] [Google Scholar]
  • 6.Sherlock R, Anderson P, Doyle L, for the Victorian Infant Collaborative Study Group (VICS) Neurodevelopmental sequelae of intraventricular haemorrhage at 8 years of age in a regional cohort of ELBW/very preterm infants. Early Human Development. 2005;81:909–16. doi: 10.1016/j.earlhumdev.2005.07.007. [DOI] [PubMed] [Google Scholar]
  • 7.Cheong JL, Anderson P, Roberts G, Duff J, Doyle LW, for the Victorian Infant Collaborative Study Group (VICS) Postnatal corticosteroids and neurodevelopmental outcomes in extremely low birthweight or extremely preterm infants: 15-year experience in Victoria, Australia. Archives of Disease in Childhood-Fetal and Neonatal Edition. 2013;98:F32–F6. doi: 10.1136/fetalneonatal-2011-301355. [DOI] [PubMed] [Google Scholar]
  • 8.Filan PM, Hunt RW, Anderson PJ, Doyle LW, Inder TE. Neurologic outcomes in very preterm infants undergoing surgery. Journal of Pediatrics. 2012;160:409–14. doi: 10.1016/j.jpeds.2011.09.009. [DOI] [PubMed] [Google Scholar]
  • 9.Ment LR, Hirtz D, Huppi PS. Imaging biomarkers of outcome in the developing preterm brain. Lancet Neurology. 2009;8:1042–55. doi: 10.1016/S1474-4422(09)70257-1. [DOI] [PubMed] [Google Scholar]
  • 10.Miller SP, Ferriero DM. From selective vulnerability to connectivity: insights from newborn brain imaging. Trends in Neurosciences. 2009;32:496–505. doi: 10.1016/j.tins.2009.05.010. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Volpe JJ. Brain injury in premature infants: a complex amalgam of destructive and developmental disturbances. Lancet Neurology. 2009;8:110–24. doi: 10.1016/S1474-4422(08)70294-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Inder TE, Wells SJ, Mogridge NB, Spencer C, Volpe JJ. Defining the nature of the cerebral abnormalities in the premature infant: A qualitative magnetic resonance imaging study. Journal of Pediatrics. 2003;143:171–9. doi: 10.1067/S0022-3476(03)00357-3. [DOI] [PubMed] [Google Scholar]
  • 13.Miller SP, Ferriero DM, Leonard C, Piecuch R, Glidden DV, Partridge JC, et al. Early brain injury in premature newborns detected with magnetic resonance imaging is associated with adverse early neurodevelopmental outcome. Journal of Pediatrics. 2005;147:609–16. doi: 10.1016/j.jpeds.2005.06.033. [DOI] [PubMed] [Google Scholar]
  • 14.Woodward LJ, Anderson PJ, Austin NC, Howard K, Inder TE. Neonatal MRI to predict neurodevelopmental outcomes in preterm infants. New England Journal of Medicine. 2006;355:685–94. doi: 10.1056/NEJMoa053792. [DOI] [PubMed] [Google Scholar]
  • 15.Iwata S, Nakamura T, Hizume E, Kihara H, Takashima S, Matsuishi T, et al. Qualitative brain MRI at term and cognitive outcomes at 9 years after very preterm birth. Pediatrics. 2012;129:E1138–E47. doi: 10.1542/peds.2011-1735. [DOI] [PubMed] [Google Scholar]
  • 16.Woodward LJ, Clark CAC, Bora S, Inder TE. Neonatal white matter abnormalities an important predictor of neurocognitive outcome for very preterm children. PLOS One. 2012;7:e51879. doi: 10.1371/journal.pone.0051879. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Kidokoro H, Neil JJ, Inder TE. New MR imaging assessment tool to define brain abnormalities in very preterm infants at term. American Journal of Neuroradiology. 2013;34:2208–14. doi: 10.3174/ajnr.A3521. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Murray AL, Scratch SE, Thompson DK, Inder TE, Doyle LW, Anderson JFI, et al. Neonatal brain pathology predicts adverse attention and processing speed outcomes in very preterm and/or very low birth weight children. Neuropsychology. 2014;28:552–62. doi: 10.1037/neu0000071. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Omizzolo C, Scratch SE, Stargatt R, Kidokoro H, Thompson DK, Lee KJ, et al. Neonatal brain abnormalities and memory and learning outcomes at 7 years in children born very preterm. Memory. 2014;22:605–15. doi: 10.1080/09658211.2013.809765. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Reidy N, Morgan A, Thompson DK, Inder TE, Doyle LW, Anderson PJ. Impaired language abilities and white matter abnormalities in children born very preterm and/or very low birth weight. Journal of Pediatrics. 2013;162:719–24. doi: 10.1016/j.jpeds.2012.10.017. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.Tich SNT, Anderson PJ, Hunt RW, Lee KJ, Doyle LW, Inder TE. Neurodevelopmental and perinatal correlates of simple brain metrics in very preterm infants. Archives of Pediatrics & Adolescent Medicine. 2011;165:216–22. doi: 10.1001/archpediatrics.2011.9. [DOI] [PubMed] [Google Scholar]
  • 22.Wechsler D. Wechsler abbreviated scale of intelligence. New York: Psychological Corporation; 1989. [Google Scholar]
  • 23.Wilkinson GS, Robertson GJ. WRAT 4: Wide Range Achievement Test – professional manual. Lutz, Florida: Psychological Assessment Resources; 2006. [Google Scholar]
  • 24.Henderson SE, Sugden D, Barnett AL. Movement Assessment Battery for Children—second edition (Movement ABC-2) London, UK: The Psychological Corporation; 2007. [Google Scholar]
  • 25.Goodman R. Psychometric properties of the Strengths and Difficulties Questionnaire. Journal of the American Academy of Child and Adolescent Psychiatry. 2001;40:1337–45. doi: 10.1097/00004583-200111000-00015. [DOI] [PubMed] [Google Scholar]
  • 26.Wilson-Ching M, Pascoe L, Doyle LW, Anderson PJ. Effects of correcting for prematurity on cognitive test scores in childhood. Journal of Paediatrics and Child Health. 2014;50:182–8. doi: 10.1111/jpc.12475. [DOI] [PubMed] [Google Scholar]
  • 27.Cole TJ, Freeman JV, Preece MA. British 1990 growth reference centiles for weight, height, body mass index and head circumference fitted by maximum penalized likelihood. Statistics in Medicine. 1998;17:407–29. [PubMed] [Google Scholar]
  • 28.Shah DK, Doyle LW, Anderson PJ, Bear M, Daley AJ, Hunt RW, et al. Adverse neurodevelopment in preterm infants with postnatal sepsis or necrotizing enterocolitis is mediated by white matter abnormalities on magnetic resonance imaging at term. Journal of Pediatrics. 2008;153:170–5. doi: 10.1016/j.jpeds.2008.02.033. [DOI] [PubMed] [Google Scholar]
  • 29.Roberts G, Howard K, Spittle AJ, Brown NC, Anderson PJ, Doyle LW. Rates of early intervention services in very preterm children with developmental disabilities at age 2 years. Journal of Paediatrics and Child Health. 2008;44:276–80. doi: 10.1111/j.1440-1754.2007.01251.x. [DOI] [PubMed] [Google Scholar]
  • 30.StataCorp. Stata Statistical Software: Release 14. College Station, TX: StataCorp LP; 2015. [Google Scholar]
  • 31.Carlin JB, Gurrin LC, Sterne JAC, Morley R, Dwyer T. Regression models for twin studies: A critical review. International Journal of Epidemiology. 2005;34:1089–99. doi: 10.1093/ije/dyi153. [DOI] [PubMed] [Google Scholar]
  • 32.Hanley JA, Negassa A, Edwardes MDD, Forrester JE. Statistical analysis of correlated data using generalized estimating equations: An orientation. American Journal of Epidemiology. 2003;157:364–75. doi: 10.1093/aje/kwf215. [DOI] [PubMed] [Google Scholar]
  • 33.Stern J, Smith G. Sifting the evidence-what’s wrong with significance tests? British Medical Journal. 2001;322:226–31. doi: 10.1136/bmj.322.7280.226. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34.Spittle AJ, Cheong J, Doyle LW, Roberts G, Lee KJ, Lim J, et al. Neonatal white matter abnormality predicts childhood motor impairment in very preterm children. Developmental Medicine and Child Neurology. 2011;53:1000–6. doi: 10.1111/j.1469-8749.2011.04095.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35.Volpe JJ. Cerebellum of the premature infant: rapidly developing, vulnerable, clinically important. Journal of Child Neurology. 2009;24:1085–104. doi: 10.1177/0883073809338067. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36.Alexander GE, Delong MR, Strick PL. Parallel organization of functionally segregated circuits linking basal ganglia and cortex. Annual Review of Neuroscience. 1986;9:357–81. doi: 10.1146/annurev.ne.09.030186.002041. [DOI] [PubMed] [Google Scholar]
  • 37.Fama R, Sullivan EV. Thalamic structures and associated cognitive functions: Relations with age and aging. Neuroscience and Biobehavioral Reviews. 2015;54:29–37. doi: 10.1016/j.neubiorev.2015.03.008. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38.Schmahmann JD, Sherman JC. The cerebellar cognitive affective syndrome. Brain. 1998;121:561–79. doi: 10.1093/brain/121.4.561. [DOI] [PubMed] [Google Scholar]
  • 39.Steiner H, Tseng KY. Handbook of basal ganglia structure and function. New York: Academic Press; 2010. [Google Scholar]
  • 40.Pierson CR, Folkerth RD, Billiards SS, Trachtenberg FL, Drinkwater ME, Volpe JJ, et al. Gray matter injury associated with periventricular leukomalacia in the premature infant. Acta Neuropathologica. 2007;114:619–31. doi: 10.1007/s00401-007-0295-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 41.Limperopoulos C, Chilingaryan G, Sullivan N, Guizard N, Robertson RL, du Plessis AJ. Injury to the premature cerebellum: Outcome is related to remote cortical development. Cerebral Cortex. 2014;24:728–36. doi: 10.1093/cercor/bhs354. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 42.Makropoulos A, Aljabar P, Wright R, Huning B, Merchant N, Arichi T, et al. Regional growth and atlasing of the developing human brain. Neuroimage. 2016;125:456–78. doi: 10.1016/j.neuroimage.2015.10.047. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 43.Abernethy LJ, Cooke RW, Foulder-Hughes L. Caudate and hippocampal volumes, intelligence, and motor impairment in 7-year-old children who were born preterm. Pediatr Res. 2004;55:884–93. doi: 10.1203/01.PDR.0000117843.21534.49. [DOI] [PubMed] [Google Scholar]
  • 44.Kesler SR, Ment LR, Vohr B, Pajot SK, Schneider KC, Katz KH, et al. Volumetric analysis of regional cerebral development in preterm children. Pediatr Neurol. 2004;31:318–25. doi: 10.1016/j.pediatrneurol.2004.06.008. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 45.Peterson BS, Vohr B, Staib LH, Cannistraci CJ, Dolberg A, Schneider KC, et al. Regional brain volume abnormalities and long-term cognitive outcome in preterm infants. JAMA. 2000;284:1939–47. doi: 10.1001/jama.284.15.1939. [DOI] [PubMed] [Google Scholar]
  • 46.Cheong JL, Anderson PJ, Roberts G, Burnett AC, Lee KJ, Thompson DK, et al. Contribution of brain size to IQ and educational underperformance in extremely preterm adolescents. PLoS One. 2013;8:e77475. doi: 10.1371/journal.pone.0077475. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 47.Hoon AH, Jr, Stashinko EE, Nagae LM, Lin DD, Keller J, Bastian A, et al. Sensory and motor deficits in children with cerebral palsy born preterm correlate with diffusion tensor imaging abnormalities in thalamocortical pathways. Dev Med Child Neurol. 2009;51:697–704. doi: 10.1111/j.1469-8749.2009.03306.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 48.Setanen S, Lehtonen L, Parkkola R, Aho K, Haataja L, on behalf of the PIPARI Study Group Prediction of neuromotor outcome in infants born preterm at 11 years of age using volumetric neonatal magnetic resonance imaging and neurological examinations. Dev Med Child Neurol. 2016;58:721–7. doi: 10.1111/dmcn.13030. [DOI] [PubMed] [Google Scholar]
  • 49.Doyle L, Cheong J, Burnett A, Roberts G, Lee K, Anderson P, et al. Biological and social influences on outcomes of extreme-preterm/low-birth-weight adolescents. Pediatrics. 2015;136:e1513–20. doi: 10.1542/peds.2015-2006. [DOI] [PubMed] [Google Scholar]
  • 50.De Vries LS, Van Haastert ILC, Rademaker KJ, Koopman C, Groenedaal F. Ultrasound abnormalities preceding cerebral palsy in high-risk preterm infants. Journal of Pediatrics. 2004;144:815–20. doi: 10.1016/j.jpeds.2004.03.034. [DOI] [PubMed] [Google Scholar]
  • 51.de Vries LS, Volpe JJ. Value of sequential MRI in preterm infants. Neurology. 2013;81:2062–3. doi: 10.1212/01.wnl.0000437309.22603.0a. [DOI] [PubMed] [Google Scholar]

RESOURCES