Search

UNLABELLED The aim of this investigation was to verify whether plasma S100B could be a useful tool in identifying which infants subjected to extracorporeal membrane oxygenation (ECMO) might develop intracranial haemorrhage (ICH). A case-control study of eight infants who developed ICH during ECMO was conducted. Plasma samples collected daily after ECMO insertion were assessed for S100B and compared with those obtained from eight infants supported by ECMO who did not develop ICH. Cerebral ultrasound and Doppler velocimetry waveform patterns in the middle cerebral artery (MCA PI) were also recorded at the same time as blood sampling. S100B blood concentrations were significantly higher in the group of infants with ICH 72 h before any signs of haemorrhage could be detected by ultrasound (ICH: 2.91 +/- 0.91 microg/L vs. control: 0.53 +/- 0.15 microg/L), reaching their peak at day 6, when cerebral ultrasound scan patterns were suggestive of intracranial haemorrhage (ICH: 3.50 +/- 1.03 microg/L vs. control: 0.66 +/- 0.27 microg/L) (p < 0.05, for both). The highest S100B levels were observed in the three ICH infants who expired during the ECMO procedure (3.43 microg/L, 4.0 microg/L, 4.12 microg/L, respectively). MCA PI values in the ICH group were also significantly higher, but only 24 h before any ultrasound pattern of bleeding was detected (ICH: 2.31 +/- 0.22 vs control: 1.81 +/- 0.24) (p < 0.05). CONCLUSION This study suggests that blood S100B measurement could be a promising tool for the identification of infants at risk of ICH when imaging assessment and clinical symptoms of haemorrhage might still be silent.