Changes in cerebral blood flow (CBF) and metabolism are now widely used to map and quantify neural activity, although the underlying mechanism for these changes is still incompletely understood. Magnetic resonance spectroscopy (MRS) at 3T, synchronized with a 32-s block design visual stimulation paradigm, was employed to investigate activation-induced changes in temperature and metabolism in the human primary visual cortex. A marginally significant increase in the local temperature of the visual cortex was found (0.1°C, P ؍ 0.09), excluding the possibility of a temperature decrease (95% confidence interval (CI) ؍ 0.0 -0.2°C), which was previously suggested. A comparison with models of thermal equilibrium in the presence of blood flow suggests that an increase in heat production during activation, greater than or at least equal to that produced by the complete oxidative metabolism of the elevated glucose (Glc) utilization accompanying activation, would be required to offset the cooling effects of the increased blood flow. In activated regions of the brain, blood flow and glucose (Glc) consumption increase substantially more than oxygen consumption (1). These changes, along with changes in cerebral blood volume (CBV), lead to the blood oxygenation level-dependent contrast (BOLD) in functional magnetic resonance imaging (fMRI) (1). Both cerebral blood flow (CBF) and metabolism are involved in brain thermoregulation, and pharmacological changes of these processes lead to changes in brain temperature (2). Broad investigations have been concerned with the possible mechanisms that underlie the BOLD effect and its relation to neuronal activation (3). Magnetic resonance spectroscopy (MRS) provides a noninvasive tool to investigate these processes in the conscious human brain, because MRS properties, such as signal area, frequency shift, and line width, can serve as probes of metabolite concentration, temperature, and oxygenation level.Previous MRS studies used prolonged stimulation duration (4 -48 min) (4 -9). However, long slow changes are prone to systematic errors. The contribution of drifts to systematic errors can be diminished with the use of a block design paradigm. We chose multiple 32-s blocks as an optimal timing to follow the BOLD response, oxidative Glc metabolism, and anaerobic glycolysis (3,10). The same block design behavioral paradigm was used for both fMRI and MRS studies (at 3T) in order to explore the physiologic and metabolic correlates of brain activation.
MATERIALS AND METHODS
SubjectsNine subjects (five men and four women, 28 -57 years of age) underwent MRI, fMRI, and 1 H-MRS of the brain. Informed consent was obtained in accordance with the guidelines of the Institutional Review Board of the Beth Israel Deaconess Medical Center.
MRI and Visual StimulationThe studies were performed on a 3T scanner (Signa VH/i, GE Healthcare, Waukesha, WI) equipped with a transmit body coil and a receive-only 7.8-cm surface coil. The subjects entered the scanner in a supine position, and the calcarine fissure wa...