Article Text

Download PDFPDF

Serum S-100 protein does not correlate with cerebral ultrasound scans in preterm infants
  1. T Chant,
  2. J May,
  3. A J B Emmerson
  1. Neonatal Medical Unit, St Mary’s Hospital, Whitworth Park, Manchester M13 0JK, UK; anthony.emmerson{at}

Statistics from

Whitelaw et al1 recently reported that S-100 protein concentration was 20–200 times higher than control levels in the cerebrospinal fluid of infants with posthaemorrhagic ventricular dilatation. S-100 protein is produced only within the brain by astrocytes, but it can be detected in the serum after cerebral damage in adults with stroke, where it is a marker of infarction volume,2 and it has also been found to be useful in head injury.3 A study of term infants found that serum S-100 protein levels were detectable after uncomplicated delivery.4 Cerebrospinal fluid is not often taken from preterm infants, and an easily obtainable serum marker for brain cell damage would be of value as an indicator of the degree of cerebral insult in such infants.

The aims of our study were to determine firstly whether S-100 protein could be measured in the serum of preterm infants during the first week after birth, and secondly whether there was a significant difference in these levels between infants who had and had not suffered parenchymal damage as diagnosed by cerebral ultrasound scans taken during the first week after birth.

Thirty preterm infants of 25–35 weeks gestation were recruited after signed parental consent, and 0.5 ml blood samples for S-100 determination were taken at the same time as routine phlebotomy on days 1, 2, 3, 5, and 7 after birth. Routine cerebral ultrasound scans were undertaken during the stay on the unit. The study was approved by the Manchester research ethics committee (central).

S-100 protein was detectable in the serum of preterm infants and concentrations ranged from 0.85 to 22.0 μg/l (table 1).

Seven of the 30 infants had features of parenchymal damage on cranial ultrasound scan, with a median S-100 protein level of 3.39 μg/l. The remaining 23 infants had normal scans and had a median S100 protein level of 3.18 μg/l (table 2). The data were analysed using a Mann-Whitney U test, and no significant difference in levels of S100 protein was found between the two groups (p = 0.774).

This study confirmed the presence of S-100 protein in the serum of preterm infants, but, in view of these findings, its measurement would not be useful as a marker of cerebral damage.

Acknowledgements: Cambridge Life Sciences performed the S-100 protein analyses.

Table 1

Serum S-100 protein levels in preterm infants of 25–35 weeks gestation by day after birth

Table 2

Serum S-100 protein levels in preterm infants with and without parenchymal damage as shown on cerebral ultrasound scanning


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.