Background: Furosemide and acetazolamide are widely used in the treatment of posthaemorrhagic ventricular dilatation (PHVD) in the hope of avoiding the need for surgical management, but this approach has not been evaluated in a controlled trial. This multicentre randomised controlled trial tested the hypothesis that these drugs would reduce the rate of shunt placement and increase disability-free survival at 1 year of age.
Methods: Between 1992 and 1996, 177 infants aged less than 3 months past term, and with ventricular width more than 4 mm above 97th centile after intraventricular haemorrhage, were randomly assigned standard therapy alone or standard therapy plus treatment with acetazolamide (100 mg/kg daily) and furosemide (1 mg/kg daily). A minimisation algorithm ensured balance between groups with respect to both referral centre and the presence of a cerebral parenchymal lesion on cerebral ultrasonography at enrolment. The trial was stopped in September, 1996, because the data showed a clear advantage with standard therapy.
Findings: We report outcomes for 151 infants whose expected date of delivery was before the end of 1995, with complete information at 1 year for 129 infants. The median gestational age was 28 weeks, mean birthweight 1299 g, and mean postnatal age at enrolment 25 days. 44% had a parenchymal lesion at randomisation. Death or shunt placement occurred in 49 of 75 infants allocated drugs plus standard therapy, compared with 35 of 76 allocated to standard therapy alone. The relative risk was 1.42 (95% CI 1.06-1.90; p=0.026), which is equivalent to one extra death or shunt placement for every five infants allocated drug therapy. 84% (52/62) of infants assigned drug therapy had died or were disabled or impaired at 1 year, compared with 60% (40/67) of those assigned standard therapy (relative risk 1.40 [1.12-1.76]; p=0.012).
Interpretation: These preliminary results suggest that the use of acetazolamide and furosemide in preterm infants with PHVD is associated with a higher rate of shunt placement and increased neurological morbidity, and so cannot be recommended.