Abstract

One hundred and fifty seven infants with progressive ventricular dilatation after intraventricular haemorrhage were randomised to either early repeated cerebrospinal fluid tapping or conservative management. Thirty two (20%) infants died and 13 (8%) were lost to follow up. One hundred and twelve children (90% of survivors) were examined at 30 months by a single experienced examiner. Overall, 54 (48%) scored less than 70 on the Griffiths developmental scales, 101 (90%) had neuromotor impairment, and 85 (76%) had marked disability; 63 (56%) had multiple impairments. Vision was severely affected in 10 (9%) and 30 (27%) had a field defect. Six per cent (seven children) had sensorineural hearing loss and 16 (14%) were taking regular anticonvulsant drugs. Although early cerebrospinal fluid tapping reduced the rate of ventricular and head expansion, there was no statistically significant difference (at the 5% level) between the treatment groups in the prevalence of neuromotor impairments, non-neuromotor impairments, nor multiple impairments at 30 months. These findings were consistent regardless of the presence or absence of a parenchymal cerebral lesion at entry to the trial. In the light of these findings and the 7% risk of cerebrospinal fluid infection associated with repeated tapping, this form of early intervention cannot be recommended.