Objective To determine whether maternal allopurinol treatment during suspected fetal hypoxia would reduce the release of biomarkers associated with neonatal brain damage.

Design A randomised double-blind placebo controlled multicentre trial.

Patients We studied women in labour at term with clinical indices of fetal hypoxia, prompting immediate delivery.

Setting Delivery rooms of 11 Dutch hospitals.

Intervention When immediate delivery was foreseen based on suspected fetal hypoxia, women were allocated to receive allopurinol 500 mg intravenous (ALLO) or placebo intravenous (CONT).

Main outcome measures Primary endpoint was the difference in cord S100ß, a tissue-specific biomarker for brain damage.

Results 222 women were randomised to receive allopurinol (ALLO, n=111) or placebo (CONT, n=111). Cord S100ß was not significantly different between the two groups: 44.5 pg/mL (IQR 20.2–71.4) in the ALLO group versus 54.9 pg/mL (IQR 26.8–94.7) in the CONT group (difference in median −7.69 (95% CI −24.9 to 9.52)). Post hoc subgroup analysis showed a potential treatment effect of allopurinol on the proportion of infants with a cord S100ß value above the 75th percentile in girls (ALLO n=5 (12%) vs CONT n=10 (31%); risk ratio (RR) 0.37 (95% CI 0.14 to 0.99)) but not in boys (ALLO n=18 (32%) vs CONT n=15 (25%); RR 1.4 (95% CI 0.84 to 2.3)). Also, cord neuroketal levels were significantly lower in girls treated with allopurinol as compared with placebo treated girls: 18.0 pg/mL (95% CI 12.1 to 26.9) in the ALLO group versus 32.2 pg/mL (95% CI 22.7 to 45.7) in the CONT group (geometric mean difference −16.4 (95% CI −24.6 to −1.64)).

Conclusions Maternal treatment with allopurinol during fetal hypoxia did not significantly lower neuronal damage markers in cord blood. Post hoc analysis revealed a potential beneficial treatment effect in girls.

Trial registration number NCT00189007, Dutch Trial Register NTR1383.