 |
|
 |
|
 |
 |
|
 |
 |
 |
 |
 |
 |
 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
PEDIATRICS Vol. 106 No. 2 August 2000, pp. 235-243
,
From the Department of * Paediatrics and Child Health and
Robert Steiner Magnetic Resonance Unit, Imperial College of Science,
Technology and Medicine, Hammersmith Campus, London, United
Kingdom; and § Department of Paediatric Neurology and
Psychiatry, Catholic University, Rome, Italy.
 |
ABSTRACT |
|---|
Top
Abstract
Methods
Results
Discussion
Conclusion
References
|
|---|
Objectives. The aims of the study were to establish the relationship between head growth in the first year of life with the pattern on injury on neonatal magnetic resonance imaging (MRI) in infants with hypoxic-ischemic encephalopathy (HIE) and to relate these to the neurodevelopmental outcome.
Methods. Fifty-two term infants who presented at birth with a neonatal encephalopathy consistent with HIE and who had neonatal brain MRI were entered into the study. Head circumference charts were evaluated retrospectively and the head growth over the first year of life compared with the pattern of brain lesions on MRI and with the neurodevelopmental outcome at 1 year of age. Suboptimal head growth was classified as a drop of >2 standard deviations across the percentiles with or without the development of microcephaly, which was classified as a head circumference below the third percentile.
Results. There was no statistical difference between the neonatal head circumferences of the infants presenting with HIE and control infants. At 12 months, microcephaly was present in 48% of the infants with HIE, compared with 3% of the controls. Suboptimal head growth was documented in 53% of the infants with HIE, compared with 3% of the controls. Suboptimal head growth was significantly associated with the pattern of brain lesions, in particular to involvement of severe white matter and to severe basal ganglia and thalamic lesions. Suboptimal head growth predicted abnormal neurodevelopmental outcome with a sensitivity of 79% and a specificity of 78%, compared with the presence of microcephaly at 1 year of age, which had a sensitivity of only 65% and a specificity of 73%. The exceptions were explained by infants with only moderate white matter abnormalities who had suboptimal head growth but normal outcome at 1 year of age and by infants with moderate basal ganglia and thalamic lesions only who had normal head growth but significant motor abnormality. Key words: hypoxic-ischemic encephalopathy, brain, neonate, head circumference, magnetic resonance imaging, neurodevelopmental outcome.
Hypoxic-ischemic encephalopathy (HIE) occurs in 1.5 to 6 per 1000 live births.1 The advent of neonatal magnetic
resonance imaging (MRI) has allowed the identification of a spectrum of
lesions in these patients ranging from minimal localized changes to
diffuse abnormalities in the basal ganglia and thalami, white matter,
and cortex.2-4
MRI has become a valuable tool for predicting outcome in infants with
HIE in the neonatal period.5,6 It is well-recognized that
some infants with signs of HIE develop a secondary microcephaly. These
children usually have a normal head circumference at birth but the
normal rate of head growth is not maintained in the first months of
life. Not all infants who sustain a hypoxic-ischemic insult, however,
develop microcephaly,7 and there are no published studies
relating the variability in head growth with the pattern of brain
injury after HIE.
The aim of this study was to compare the pattern of initial brain
injury on MRI with the head growth in the first year of life in a
cohort of infants with HIE. More specifically, we wished to evaluate:
1) the relationship between head growth and MRI findings in the
neonatal period, and 2) the relationship between head growth and
neurodevelopmental outcome.
Ethical permission for this study was obtained from the
Hammersmith Hospital Research Ethics Committee. The infants described in this study are part of a large prospective cohort of term infants born since October 1991 with perinatal hypoxic-ischemic brain injury,
who were born at or referred to the Hammersmith Hospital, London for
MRI. The diagnosis of HIE was made in infants who showed signs of fetal
distress before delivery, who had abnormal Apgar scores (<5 at 1 minute and <8 at 5 minutes), requiring resuscitation at birth, and who
developed neurological abnormalities during the first 24 hours after
delivery. Signs of fetal distress included abnormal cardiotocograph
recordings, such as decreased variability, late decelerations, and a
baseline bradycardia (<100/minute) with or without meconium stained
liquor. Neurological abnormalities included abnormal tone, poor
feeding, convulsions, and altered conscious level. HIE was classified
during the first week of life as mild, moderate, or severe (stages 1, 2, or 3 according to Sarnat and Sarnat8) by the attending
neonatologist.
Term infants (>37 weeks of gestation) were included in the study if
they fulfilled our criteria for the diagnosis of HIE and they had: 1)
at least 1 brain MRI performed in the neonatal period between 1 and 4 weeks from delivery, 2) a head circumference measured within the first
week of life and at 12 months of age, and 3) detailed
neurodevelopmental follow-up performed in this hospital. Infants were
born between October 1991 and January 1998.
Infants who were subsequently diagnosed as suffering from genetic or
metabolic syndromes or who presented with other neonatal complications,
such as neonatal meningitis, were excluded from the study. Infants with
preexisting abnormalities on MRI suggesting antenatal insult or
congenital malformation were also excluded from the study.
MRI
Infants were scanned using a 1 Tesla HPQ magnet. Images were
obtained in the transverse plane with T1-weighted spin echo (SE 860/20), T2 weighted spin echo (SE 3000/120), and age-related inversion recovery (IR 3800/30/950) sequences.
Images obtained between 1 and 4 weeks of life, when the pattern of
injury is easiest to define, were assessed for abnormal signal
intensities by an experienced observer (M.A.R.) blinded to the head
circumference data. The pattern of abnormal signal intensities observed
was documented as follows:
METHODS
Top
Abstract
Methods
Results
Discussion
Conclusion
References
Cortical highlighting is usually associated with the development of abnormal signal intensity within the adjacent subcortical white matter during the second week.2 Apparent highlighting of the cortex around the central fissure may be seen in control patients. It is difficult to distinguish these mild changes from early myelination, the presence of a small amount of subarachnoid blood or partial volume effects. Grade 1 highlighting in this region was called normal unless it was associated with abnormal signal intensity within the adjacent white matter.
The scans were classified according to the predominant pattern observed into 8 groups, which are summarized in Tables 1 and 2.
|
|
Follow-Up Imaging
Follow-up imaging was performed at regular intervals over the first year of life in the majority of infants. Ventricular dilatation and widening of the extracerebral space may contribute to head growth; therefore, these 2 areas were assessed. The extracerebral space may widen during the first year of life.9,10 This is usually a benign condition and is associated with an increase in the head circumference percentile.9,10 It contrasts with widening that is secondary to atrophy of the brain, where the head circumference percentile falls or possibly remains static. Visual analysis of all slices of the brain on follow-up images at ~1 year of age was performed, and the presence of ventricular dilation and/or enlargement of the extracerebral space were documented.
Head Circumference
The maximum occipital-frontal head circumference was measured during the first week after birth before discharge from hospital and each time the infants attended the hospital for MRI or clinical follow-up. The head circumference was checked as part of the routine clinical assessment using a paper or plastic tape measure. The head circumference measurements of the infants in the study were evaluated retrospectively and plotted on the charts designed by the Child Growth Foundation 1996, using specific charts for boys and girls.11 Infants were defined as microcephalic if their head circumference fell below the third percentile on the appropriate chart. To identify infants with suboptimal head growth but head circumferences above the third percentile, the number of standard deviations (SDs) crossed on the chart over the first year were documented for each child using a computerized program.12 Suboptimal head growth was classified as a drop in percentiles of >2 SD.
Neurodevelopmental Outcome
All the children were examined at 1 year of age by an experienced neurodevelopmental pediatrician (F.M.C., L.M.D., or E.M.). Neurological examination was performed using a structured proforma,13 looking in particular at posture, muscle tone, and power and reflexes.
Developmental scales by Griffiths14 were used to evaluate neurodevelopment. The results were classified as abnormal when the developmental quotient (DQ) fell below 85.
Controls
The head circumference data from the infants with HIE were compared with the head circumferences measured at birth and 12 months of age in a normal control group of 63 term infants born within our maternity unit at Queen Charlottes Hospital. This group was used to document individual variations in head growth in normal children over the first year of life to be assessed. The control group had the same racial, gestational age, and sex distribution as our patient group and had a detailed neurological examination at birth and a neurodevelopmental follow-up at 1 year of age.12
Statistical Analysis
Nonparametric statistics (
2 test) was
used for the analysis of the results. The level of significance was set
at .01. Sensitivity and specificity of both head growth and
microcephaly in predicting outcome also were calculated.
| |
RESULTS |
|---|
|
Top
Abstract
Methods
Results
Discussion
Conclusion
References
|
|---|
Between October 1991 and January 1998, 118 term infants fulfilling our criteria for HIE were scanned in our unit. Twenty- six of these infants died within the first year of life. Twenty-nine infants were referred from outside hospitals and were not systematically followed up in our unit. One infant had a postnatal infection and developed new MRI findings at that time, 2 infants were followed but not seen at 1 year of age, and 1 infant moved abroad. Seven infants were treated with postnatal hypothermia and were not included in the study. There were no infants with preexisting antenatal lesions that excluded them from the study. Fifty-two infants (27 boys and 25 girls) fulfilled all our study criteria. For the study infants, gestational age ranged between 37.6 and 42.3 weeks and birth weight ranged from 2.23 to 4.3 kg. Fifteen infants had stage 1 HIE and 37 infants had stage 2 HIE (see Table 1).
Neurodevelopmental Follow-Up
Twenty-three of the 52 infants with HIE had a normal neurological examination, 6 infants had mild hypotonia, 1 had hemiplegia, 3 had diplegia, and 19 had quadriplegia.
Eight of the infants with quadriplegia had severe global retardation
and were untestable with developmental scales. In the remaining 44 infants, the DQ ranged from 34 to 131. Thirty-three of the 44 infants
had a normal DQ (
85) and 11 had a quotient of <85 (Table 1).
MRI
The numbers of infants in each imaging group are shown in Tables 1 and 2, and examples of each pattern of lesion are shown in Fig 1A-G.
|
Twelve of the 52 infants had normal scans by the end of the first week, although they may have shown signs of brain swelling on the scan performed in the first days of life. Four of the 9 children with severe white matter abnormalities had changes consistent with hemorrhagic infarction.
Follow-Up Imaging
Thirty-six of the 52 infants had follow-up imaging at ~1 year of age. All but 1 of the infants with microcephaly had some evidence of ventricular dilation and/or widening of the extracerebral space. The case with the most marked ventricular dilation is shown in Fig 1H. This infant had hemorrhagic white matter infarction on neonatal scans with some intraventricular hemorrhage (Fig 1E).
Head Circumference
On the first measurement, the head circumference ranged from 31.5 to 37.7 cm (mean: 35.05; SD: 1.4) in the infants with HIE and from 31 to 38 cm (mean: 34.4; SD: 1.5) in the control group. One of the 53 infants with HIE and 2 of the 63 control infants had a head circumference below the third percentile at birth.
At 12 months of age the head circumference ranged from 37.5 to 49.2 cm (mean: 44.7; SD: 2.5) in the infants with HIE and from 43.2 to 51 cm (mean: 46.6; SD: 1.5) in the control group. Twenty-five of the 52 infants (48%) with HIE became microcephalic by 12 months of age with a head circumference below the third percentile. In the control group, only 2 of the 63 (3%) were microcephalic at 1 year of age.
Head Growth
At 1 year of age, 28 of 52 infants (53%) with HIE had a head circumference of >2 SD below the birth percentile, compared with only 2 of 63 in the control group. The difference between the 2 groups was significant (P < .001).
Microcephaly and Head Growth
At 12 months old, 21 children showed suboptimal head growth and their head circumference was below the third percentile. Seven other children showed suboptimal head growth, but their head circumference at 12 months of age remained above the third percentile. Four children had a head circumference below the third percentile at 12 months of age but the changes in head circumference were <2 SD below the birth percentile.
Severity of HIE and Head Circumference at Birth and 12 Months of Age
Fifteen infants had grade 1 HIE: in 11 infants, the head circumference at 12 months of age was on the same percentile or within 2 SD from the percentile found at birth and in the remaining 4 infants, it was 2 SD below. Thirty-seven infants had grade 2 HIE: in 16 infants, the head circumference at 12 months of age was on the same percentile or within 2 SD from the percentile found at birth and in 21 infants, it was 2 SD below. The difference between the 2 groups was significant (P < .001).
MRI Findings and Head Circumference at Birth and 12 Months of Age
At birth the head circumference of the infants with HIE did not differ from the normal controls, regardless of the MRI findings (Fig 2).
|
At 12 months of age, all but 1 of the infants with severe white matter changes, with moderate white matter and moderate basal ganglia and thalamic lesions, and with severe basal ganglia and thalamic lesions had a head circumference below the third percentile (Fig 2).
Details of the correlation are given in Table 2.
MRI Findings and Head Growth
Fig 3 shows individual details of head growth in the cohort of infants with HIE, subdivided according to the MRI findings.
|
All but 1 of 12 infants with normal MRI showed optimal head growth over the first year of life, ie, that the final head circumference was on the same percentile or within 2 SD of the birth circumference.
All 10 infants with mild basal ganglia lesions showed optimal head growth.
Two of 4 infants with moderate basal ganglia lesions showed optimal head growth. The other 2 showed suboptimal head growth.
All 5 infants with moderate white matter changes showed suboptimal head growth.
One of 3 infants with moderate white matter and basal ganglia lesions showed optimal head growth. All the patients with severe white matter changes and/or severe basal ganglia lesions showed suboptimal head growth.
Follow-Up MRI Findings and Head Growth
None of the infants with normal head growth had obvious ventricular dilation and/or widening of the extracerebral space. Their normal head growth was, therefore, associated with normal brain growth.
Head Growth and Neurodevelopmental Outcome
Twenty-seven infants had normal head circumferences at 12 months of age: 17 had normal and 10 had abnormal outcome.
Twenty-five infants had microcephaly (head circumference below the third percentile) at 12 months of age: 6 had normal and 19 had abnormal outcome. Details of the correlation are shown in Fig 4.
|
Twenty-four infants had optimal head growth (head circumference at 12 months of age on the same percentile or within 2 SD of the percentile at birth): 18 had normal and 6 had abnormal outcome. None of the 6 infants with normal head growth and abnormal neurology had significant ventricular dilation or widening of the extracerebral space.
Twenty-eight infants had suboptimal head growth (head circumference at 12 months of age <2 SD from the birth percentile): 5 had normal and 23 had abnormal outcome.
The presence of microcephaly predicted an abnormal developmental outcome with a sensitivity of .65 and a specificity of .73.
A suboptimal rate of head growth predicted an abnormal neurodevelopmental outcome with a sensitivity of .79 and a specificity of .78.
| |
DISCUSSION |
|---|
|
Top
Abstract
Methods
Results
Discussion
Conclusion
References
|
|---|
MRI has been used in the past to investigate the cause of microcephaly, showing that congenital brain malformations and genetic diseases are responsible in the majority of cases.15,16 In a proportion of children, microcephaly is not present at birth but becomes evident in the first months of life, suggesting a perinatal cause. A recent study has reported that ~40% of infants with acute HIE develop microcephaly within 18 months of delivery.7 Our study is in agreement with this, 48% of the infants developing microcephaly by 12 months of age. In this study, however, we were not only interested in evaluating the number of infants falling below the third percentile but also in the rate of head growth in this population. Using a computer program12 that enabled precise measurements of SD, we found suboptimal head growth in 53% of our children. Not all infants who were below the third percentile at 12 months of age had a suboptimal head growth, because 1 was already below the third percentile at birth and 3 were already below the tenth percentile and only crossed 1 SD. Conversely, there were 8 infants who had a significant decline in head growth (>2 SD) at 12 months of age but could not be defined as microcephalic because they were still on or above the third percentile. The head growth of these infants may continue to decline after 12 months of age. This possibility has already been reported in 1 study, which showed that in 60% of infants with HIE who developed microcephaly, the head circumference had not crossed the third percentile until 18 months of age.7
Suboptimal head growth was not consistently associated with the severity of HIE, although 93% of the infants with grade 1 HIE had optimal head growth, compared with only 32% of infants with grade 2 HIE.
Using neonatal brain MRI, we have been able to demonstrate that head growth was clearly related to the pattern of lesion. White matter forms the largest component of the brain and all of our study children with severe white matter injury had suboptimal head growth and developed microcephaly. All of the children with less severe white matter changes also showed suboptimal head growth, but only 40% were microcephalic at 12 months of age.
It is of interest that in our study infants with severe basal ganglia and thalamic lesions also showed poor head growth. Although all had lesions in the white matter on the follow-up scans, these were not always extensive in the neonatal period. White matter atrophy in these infants may be secondary to disruption of thalamo-cortical connections or to a relative paucity of new white matter growth. In contrast, more discrete basal ganglia and thalamic lesions, even if associated with the involvement of the internal capsule and some subcortical white matter, were not always associated with suboptimal head growth, suggesting that in these cases the degree of primary injury was not sufficient to affect white matter growth enough to produce microcephaly.
Our results do confirm previous observations that a decline in head growth is often associated with neurodevelopmental delay.19,20,21 However, we found that a suboptimal rate of growth is a better predictor of neurodevelopmental outcome than the presence of microcephaly alone. There were exceptions to these rules that could be explained by the pattern of lesions on MRI. Four of 5 infants with suboptimal head growth but normal outcome at 1 year of age had moderate white matter changes.
Conversely, 5 of 6 children with optimal head growth but abnormal outcome had moderate focal basal ganglia and thalamic lesions. The apparently normal head growth in these infants seems to reflect normal brain growth, because none of them showed ventricular dilation and/or widening of the extracerebral space, which could have masked the effects of subnormal brain growth. Infants who have suboptimal head growth but normal development at 1 year of age need to be closely followed, because assessment at 1 year of age primarily measures motor items and cognitive, perceptual, or minor neurological signs might become obvious later in childhood.
| |
CONCLUSION |
|---|
|
Top
Abstract
Methods
Results
Discussion
Conclusion
References
|
|---|
Our results show that suboptimal head growth and secondary microcephaly are frequent in infants with HIE and are related to the pattern of lesions and, in particular, to the presence of white matter lesions and severe basal ganglia and thalamic lesions. However, there are definite exceptions that are clinically significant and head circumference cannot be used, in isolation, as a reliable predictor of outcome without knowledge of the severity and site of cerebral injury. Further longitudinal prospective studies with intermediate measurements of head circumference within the first year of life are needed to evaluate the time course of the decline in head growth in the various patterns of injury. That information, together with MRI quantification of white matter and basal ganglia and thalamic volumes may clarify the relationship between white matter atrophy and basal ganglia and thalamic lesions.
| |
ACKNOWLEDGMENTS |
|---|
We thank the Medical Research Council and Picker International for their support.
| |
FOOTNOTES |
|---|
Received for publication Dec 8, 1999; accepted Dec 8, 1999.
Reprint requests to (M.A.R.) Department of Paediatrics, Hammersmith Hospital, Du Cane Rd, London, W12 OHS, United Kingdom. E-mail: m.rutherford{at}ic.ac.uk
| |
ABBREVIATIONS |
|---|
HIE, hypoxic-ischemic encephalopathy; MRI, magnetic resonance imaging; PLIC, posterior limb of the internal capsule; SD, standard deviation; DQ, developmental quotient.
| |
REFERENCES |
|---|
|
Top
Abstract
Methods
Results
Discussion
Conclusion
References
|
|---|
This article has been cited by other articles:
 |
|
 |
|
  C. M. Burns, M. A. Rutherford, J. P. Boardman, and F. M. Cowan Patterns of Cerebral Injury and Neurodevelopmental Outcomes After Symptomatic Neonatal Hypoglycemia Pediatrics, July 1, 2008; 122(1): 65 - 74. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Okereafor, J. Allsop, S. J. Counsell, J. Fitzpatrick, D. Azzopardi, M. A. Rutherford, and F. M. Cowan Patterns of Brain Injury in Neonates Exposed to Perinatal Sentinel Events Pediatrics, May 1, 2008; 121(5): 906 - 914. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. S. Brown, T. Bottiglieri, C. A. Schaefer, C. P. Quesenberry Jr, L. Liu, M. Bresnahan, and E. S. Susser Elevated Prenatal Homocysteine Levels as a Risk Factor for Schizophrenia Arch Gen Psychiatry, January 1, 2007; 64(1): 31 - 39. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 F F Gonzalez and S P Miller Does perinatal asphyxia impair cognitive function without cerebral palsy? Arch. Dis. Child. Fetal Neonatal Ed., November 1, 2006; 91(6): F454 - F459. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Rutherford, S. Counsell, J. Allsop, J. Boardman, O. Kapellou, D. Larkman, J. Hajnal, D. Edwards, and F. Cowan Diffusion-Weighted Magnetic Resonance Imaging in Term Perinatal Brain Injury: A Comparison With Site of Lesion and Time From Birth Pediatrics, October 1, 2004; 114(4): 1004 - 1014. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. Le Strange, N. Saeed, F. M. Cowan, A. D. Edwards, and M. A. Rutherford MR Imaging Quantification of Cerebellar Growth Following Hypoxic-Ischemic Injury to the Neonatal Brain AJNR Am. J. Neuroradiol., March 1, 2004; 25(3): 463 - 468. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 F. S. Ezgu, Y. Atalay, K. Gucuyener, S. Tunc, E. Koc, E. Ergenekon, and U. Tiras Neuron-Specific Enolase Levels and Neuroimaging in Asphyxiated Term Newborns J Child Neurol, November 1, 2002; 17(11): 824 - 829. [Abstract] [PDF]
|
|
|
|
 |
|
 L. R. Ment, H. S. Bada, P. Barnes, P. E. Grant, D. Hirtz, L. A. Papile, J. Pinto-Martin, M. Rivkin, and T. L. Slovis Practice parameter: Neuroimaging of the neonate: Report of the Quality Standards Subcommittee of the American Academy of Neurology and the Practice Committee of the Child Neurology Society Neurology, June 25, 2002; 58(12): 1726 - 1738. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
||
|
||
Home | My Pediatrics | Journal Information | Current Issue | Past Issues | Subscriptions & Services | Contact Us Click here for faster international access © 2000 American Academy of Pediatrics. All rights reserved. |
||
|
||
