Article Text

Download PDFPDF

Cranial ultrasound abnormalities in full term infants in a postnatal ward: outcome at 12 and 18 months
  1. Leena Haataja,
  2. Eugenio Mercuri,
  3. Frances Cowan,
  4. Lilly Dubowitz
  1. Department of Paediatrics and Neonatal Medicine, Imperial College School of Medicine, Hammersmith Hospital, Du Cane Road, London W12 0HS, UK
  1. Dr Dubowitz


OBJECTIVE To investigate whether cranial ultrasound abnormalities found in low risk full term infants had any influence on neurodevelopmental outcome.

METHODS For 103 infants who had a neurological assessment, a cranial ultrasound examination, and for whom antenatal and perinatal data were collected within 48 hours of delivery, neurodevelopmental status was evaluated at 12 and 18 months. The results of a scored neurological examination and the Griffiths mental developmental scale were correlated with the presence and type of ultrasound abnormality found in the neonatal period.

RESULTS None of the infants with ultrasound abnormalities showed any signs of cerebral palsy or severe developmental delay. There was also no significant difference between the overall neurological and neurodevelopmental scores of the infants with normal and abnormal ultrasound findings. However, when the individual subscales of the Griffiths test were analysed, all infants with bulky choroid or intraventricular haemorrhage had normal scores in all subscales, four of eight with periventricular white matter lesions had low scores on the locomotor subscale, and three of five with asymmetrical ventricles had low scores on the performance subscale. The presence of adverse antenatal and perinatal factors did not affect the outcome in this group.

CONCLUSION Incidental ultrasound abnormality in full term neonates, in particular intraventricular haemorrhage, although common, appear to have a good prognosis. Longer follow up studies are needed to see whether some of these infants, in particular those with white matter lesions, develop dyspraxia or other minor neurological impairments at school age.

Key messages

Key messages

  • Incidental ultrasound abnormalities in full term neonates are common but the infants appear to have a good prognosis at 12 and 18 months

  • Unilateral thalamic densities and intraventricular haemorrhages not complicated by parenchymal extensions were not associated with neurological sequelae

  • ultrasound
  • neurological assessment
  • intraventricular haemorrhage
  • brain
  • white matter
  • development

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

The advent of cranial ultrasound as a routine tool in neonatology has greatly improved our knowledge of the presence and incidence of brain lesions in the newborn infant. Cranial ultrasound has been used routinely for infants at risk of neurological impairment, such as those born prematurely1-5 or who have suffered from birth asphyxia,6 ,7 but less has been reported about the range of the cranial ultrasound findings and their long term significance in low risk infants.8-12

In 1996 we performed cranial ultrasound and neurological assessments in a cohort of 177 infants regarded as normal at birth. Thirty five of the 177 infants (20%) showed some ultrasound changes. As in previous studies, we observed a significant incidence of haemorrhages (6%) and asymmetrical ventricles (6%). We also observed that an additional 8% showed periventricular white matter echogenicity, which until then had not been described in normal full term infants. Some 8% of the infants had an unusually full choroid. The presence of abnormal features on ultrasound scans was associated with deviant patterns on the neurological examination (p < 0.0001).13

The aim of this study was to evaluate the neurodevelopmental outcome in this population to determine whether infants with cranial ultrasound abnormalities differ from those with normal scans.

Subjects and methods

The infants described in this study are part of a project investigating neurodevelopmental outcome in a cohort of infants admitted soon after birth to the postnatal ward at Queen Charlotte and Chelsea Hospital and regarded as normal by the obstetric and paediatric staff.14 The project was approved by the research ethics committee of the Royal Postgraduate Medical School.


While they were inpatients, mothers were asked if a neurological examination and cranial ultrasound scan could be performed on their infant within the first 48 hours of birth; 177 agreed to both.13

When the cohort of 177 infants approached the age of 12 or 18 months, their parents were contacted again to see if they were prepared to let their child have a neurological and developmental examination. Thirty seven families could not be traced because the address and the contact numbers available in the notes were no longer appropriate; three families had moved abroad, 24 had moved outside the area, and 10 were not interested in participating in the study. This resulted in a final cohort of 103 infants. Seventy six infants (46 boys and 30 girls) were assessed at the age of 12 months (range 11.7–13.3, mean 12.3), and 27 (12 boys and 15 girls) were assessed at the age of 18 months (range 17–20.3, mean 18.3).

Details of antenatal and perinatal data were available for all the children.


Cranial ultrasound scans were performed with an Advanced Technology Laboratory (ATL) mark IV sector scanner, using 5 and 7.5 MHz probes. The scans were reviewed for normal anatomy, ventricular size, and evidence of focal or diffuse increased echogenicity in the cerebral hemispheres and basal ganglia in the neonatal period.13


All the maternal notes were searched for the presence of adverse antenatal and perinatal factors to determine their influence on outcome in children with normal and abnormal ultrasound findings. Table1 shows a list of the individual antenatal and perinatal factors found.

Table 1

Individual antenatal and perinatal factors


Neurological examination

The proforma used to record the neurological findings at 12 and 18 months consists of 37 items divided into three sections.14The first section includes 26 items assessing cranial nerve function, posture, movements, tone, and reflexes. The second section includes eight items documenting developmental progress, and the third section includes three items evaluating state of behaviour. This examination has been standardised in a normal population at 12 and 18 months of age. Scores between 73 and 78 are considered normal.

Developmental assessment

The neurodevelopmental outcome was evaluated by the Griffiths mental developmental scale.15 The developmental quotients (DQs) were calculated for global development as well as for the individual subscales (locomotor, personal-social scale, hearing and language, eye and hand coordination, and performance). Results were classified as normal when the DQ was 80 or above, and abnormal when it was below 80.16

The examiner (LH) who performed the follow up assessment was not involved with the neonatal examination and was unaware of the early findings.


The results of the outcome measures were analysed in the groups with normal and abnormal ultrasound findings (Student'st test, level of significance p < 0.05). The eight children who showed an unusually full choroid were classified separately, as we found it difficult to decide whether this finding was normal or represented a choroidal haemorrhage.13


Tables 2-5 show details of the results. The incidence of abnormal ultrasound and antenatal and perinatal factors in the cohort of 103 children who participated in the follow up study was not significantly different from the incidence of the same factors in the original cohort of 177 who had a neonatal cranial ultrasound examination (p > 0.05).


A completely normal scan at birth was found in 72 of the 103 infants examined at follow up; the remaining 31 had some changes. Periventricular white matter echogenicities were present in eight of the 31, intraventricular haemorrhage in six, asymmetrical ventricles in five, unilateral thalamic densities in two, a choroid cyst in one, and a large infracerebellar space (possible arachnoid cyst) in one. An unusually full choroid was found in eight infants.


Adverse antenatal factors were found in 29 of 72 infants (40%) with a normal ultrasound and in 16 of 23 infants (61%) with abnormal ultrasound. There was no significant difference in the relative incidence of the individual factors, but intrauterine growth retardation was more common in the infants with abnormal ultrasound (5/23 (22%) v 1/72 (1%)).

Adverse perinatal factors were recorded in 11 of 72 infants (15%) with normal ultrasound and in three of 23 (13%) with abnormal ultrasound findings. There was no significant difference in the total incidence of adverse perinatal factors in the two groups, but cardiotocography abnormalities only occurred in the group with normal ultrasound. Tables2-4 show details of the type and incidence of the adverse antenatal and perinatal factors.


Neurological examination was normal in 96 and abnormal in seven of the 103 infants studied.

Of the 72 infants who had a normal cranial ultrasound scan, 66 had a normal and six an abnormal neurological examination. Of the 23 infants with abnormal ultrasound findings, 22 had a normal and one an abnormal neurological examination. All eight children with full choroid had a normal neurological examination.

When the scores on the neurological examination were analysed, the difference between the three groups was not significant: 75.93 (2.52), 76.63 (1.69), and 75.35 (2.01) for the group with normal cranial ultrasound results, the group with full choroid, and the group with abnormal ultrasound results respectively (mean (SD)). Tables 2-4 give the individual scores of the neurological examinations in the cohort in relation to ultrasound findings.


Of the 103 infants examined, 100 had normal and three had abnormal global DQ.

Eighty three of the 103 infants assessed had normal scores on all the subscales. The other 20 had mild delay (DQ between 61 and 79) on at least one subscale. None of the infants in our cohort had severe delay.

Sixty of the 72 infants with normal ultrasound scans had a normal score on all the subscales and 12 had at least one abnormal score.

Fifteen of the 23 infants with abnormal ultrasound scans had normal and eight abnormal scores on at least one of the subscales.

All eight infants with full choroid had normal scores on all the subscales.

Table 5 shows the incidence of normal and abnormal neurological examination and Griffiths scores in the cohort subdivided according to ultrasound findings.

Table 5

The incidence of normal and abnormal neurological examination and Griffiths scores in the children with different ultrasound findings

Tables 2-4 give details of the findings in all the infants. The results show that, whereas the presence of haemorrhages or full choroids was not associated with abnormal results on any of the subscales, white matter lesions were associated with locomotor delay in four of eight infants and asymmetrical ventricles with delay in performance in three of five infants (table 2). The presence of adverse antenatal or perinatal factors had no impact on outcome in infants with abnormal scans.

Table 2

Details of antenatal and perinatal factors, global neuroscore, global developmental quotients, and subscale developmental quotients in infants with abnormal ultrasound

Table 3

Details of antenatal and perinatal factors, global neuroscore, global developmental quotients and subscale developmental quotients in infants with normal ultrasound and normal outcome (n = 72)

Table 4

Details of antenatal and perinatal factors, global neuroscore, global developmental quotients and subscale developmental quotients in infants with normal ultrasound and abnormal outcome


The aim of this study was to evaluate whether the outcome of a cohort of infants regarded as normal in the neonatal period, but with abnormal features on neonatal cranial ultrasound examination, differed from those with normal scans. We were able to follow 60% of the original cohort with neonatal ultrasound and we feel there was no bias, as there was no selection other than not being able to find the infants. The refusal rate among those traced was low, and in fact the incidence of ultrasound abnormalities and adverse antenatal/perinatal factors was similar in the original and in the follow up cohort.

At follow up at 12 or 18 months, none of the children in our cohort showed any sign of cerebral palsy and there was no significant difference between the infants with and without abnormal ultrasound results on neurological examination.

Although the scores on the Griffiths developmental test in the group with ultrasound abnormalities were lower than in the group with normal cranial ultrasound, the difference was not significant The numbers were too small to allow a meaningful statistical analysis of the results in relation to the different types of ultrasound findings, but it is of interest that intraventricular haemorrhages and thalamic densities were always associated with normal scores on all the subscales. Other ultrasound abnormalities were associated with some degree of impairment. Four of the eight infants with periventricular white matter echogenicity had low gross locomotor scores, while those with asymmetrical ventricles had normal locomotor scores but three of five had lower scores in the performance scales. The infants who showed a full choroid, a finding that we were uncertain how to classify on the neonatal ultrasound, not only did not show any abnormal results but, as a group, had higher scores than the infants with normal ultrasound.

Our results are in basic agreement with previous studies which showed that unilateral thalamic densities were associated with a normal outcome.17 ,18 Also in agreement with previous studies,19-21 we found that intraventricular haemorrhage, even when large, if not complicated by severe ventricular dilatation or parenchymal extension was not associated with any neurological or developmental sequelae at 12 or 18 months of age.

As all the infants in our cohort only had one scan within 48 hours, our data on periventricular densities are not easily comparable with previous studies where outcome depends on the persistence and evolution of the lesions.22 ,23

As adverse antenatal factors such as infection and bleeding tendencies have been reported to be associated with increased risk for cerebral palsy,24 we also tried to evaluate whether the variability in outcome observed in our infants could be related to the presence of adverse antenatal or perinatal factors. Although antenatal factors were more common in the infants with ultrasound abnormalities, we were unable to find any significant association between the presence or the type of any of these factors and the outcome.

Our results suggest that some abnormality on cranial ultrasound is relatively common in a population considered to be at low risk for neurological sequelae in the neonatal unit, but they are not associated with cerebral palsy or other major neurological or developmental abnormalities. As the follow up in our cohort was relatively short, we do not know whether any of these infants would develop dyspraxia or minor neurological impairment at school age, as found in preterm infants with similar lesions. It would be of interest to know whether any children who do develop minor neurological problems also had minor developmental abnormalities on the short term follow up.