Differential fMRI-BOLD signal response to apnea in humans and anesthetized rats.

Blood oxygenation level dependent (BOLD) signal intensity (SI) and regional cerebral blood flow (CBF) during a 20-s apnea stimulus in awake humans and pentobarbital-anesthetized rats were measured to assess the usefulness of apnea in estimating cerebral vasodilatory capacity for functional MRI (fMRI) experiments. Rats were ventilated with either room air or 100% O(2.) While breathing room air, apnea for 20 s increased the BOLD SI in humans but decreased it in rats. However, in rats ventilated with 100% O(2), BOLD SI increased upon apnea for 20 s. CBF measurements in rats using laser Doppler flowmetry (LDF) showed a 45% +/- 8% increase during apnea with room air ventilation, and a 10% +/- 3% increase with 100% O(2). Arterial blood oxygen saturation fell from 96% +/- 1% to 29% +/- 5%, and cerebral tissue PO(2) decreased from 15 +/- 3 mmHg to 6 +/- 2 mmHg by the end of 20-s apnea in rats breathing room air. However, with 100% O(2) respiration, apnea produced no change in the arterial blood oxygen saturation, which remained at 99%, but increased tissue PO(2) from 35 +/- 9 mmHg to 39 +/- 10 mmHg. From the results obtained in rats ventilated with room air, it is concluded that apnea induces hypoxia that results in a decrease in fMRI-BOLD signal. The signal decrease occurred despite an increase in P(a)CO(2) and CBF. This BOLD response is the opposite of that observed in humans, who presumably do not develop hypoxia within the applied apnea period. These studies highlight the importance of the choice of ventilating gas mixture on the outcome of BOLD experiments during systemic perturbations. Copyright 2002 Wiley-Liss, Inc.