Unrestricted use of oxygen in the delivery room after preterm birth has been associated with reduced cerebral blood flow (CBF) 2 h later. To further investigate residual cerebrovascular effects of transient hyperoxia, we developed a newborn rat model in which laser-Doppler flowmetry (LDF) and near-infrared spectroscopy (NIRS) were used to monitor changes in cerebral perfusion. The hypothesis to be tested was that hyperoxic exposure limits cerebral vasodilation in response to increase in carbon dioxide tension (Pco(2)). Twenty-four 3- to 5-d-old rats were kept on spontaneous breathing with doxapram under light isoflurane anesthesia, randomized into two groups, and exposed to either room air or 100% oxygen for 30 min. Then, after 15 min of stabilization in normoxia, 8% CO(2) was given for 5 min. No significant differences in CO(2) responses were observed between the two groups: mean CBF-CO(2) reactivity as measured by NIRS was 13.3 +/- 3.9 %/kPa in the normoxia-group versus 8.8 +/- 4.1 %/kPa in the hyperoxia group (NS). The oxygenation index [(HbO(2) - Hb)/2] increased by 0.67 +/- 0.17 micromol/L/kPa in the normoxia group compared with 1.18 +/- 0.19 micromol/L/kPa in the hyperoxia group (NS). Cortical perfusion, monitored by LDF, increased by 7.3 +/- 1.5 %/kPa versus 6.8 +/- 1.8 %/kPa in the normoxia and hyperoxia groups, respectively (NS). We conclude that in newborn rats the CBF-CO(2) reactivity remains intact after 30 min of oxygen exposure.