Effect of respiratory CO(2) changes on the temporal dynamics of the hemodynamic response in functional MR imaging.

Increasing end-expiratory CO(2) levels (PETCO(2)) increases the dispersion and the time of maximum of the hemodynamic response curve in human primary visual cortex. This was demonstrated using event-related multislice functional magnetic resonance imaging (fMRI) with short repetition time and 3-s flicker light stimulation. Measurements were performed at 5 different PETCO(2) levels between 20 and 60 mmHg using hyperventilation or by adding CO(2) to the inspired air. Between 30 and 60 mmHg the full-width-at-half-maximum of the hemodynamic response curve induced by visual stimulation increased nearly linearly at 130 ms per mmHg PETCO(2). Consistent with previous studies a concomitant decrease of the signal amplitude was observed at PETCO(2) values below 40 mmHg and above 50 mmHg. The relevance of these findings for the temporal resolution of fMRI and especially of event-related methods is discussed.

Interdependence of regional and global cerebral blood flow during visual stimulation: an O-15-butanol positron emission tomography study.

The authors investigated the influence of variations in global cerebral blood flow (gCBF) on regional flow changes during visual stimulation. Global flow was varied using different end-expiratory CO2 values (PETCO2) between 20 and 70 mm Hg. Visual stimulation was performed with a red LED-array flashing at 8 Hz. Blood flow was measured with 0-15-butanol, continuous arterial blood sampling, and positron emission tomography (PET). Global flow changes surpassed the published values of O-15-H2O studies, better fitting the results of the inert gas technique (gCBF at 20, 40, and 70 mm Hg PETCO2 +/- SD was 31 +/- 4, 48 +/- 13, and 160 +/- 50 mL 100 g(-1) min(-1), respectively). The relation between PETCO2 and CBF in the current study was best described by an exponential rather than a linear function. At low PETCO2, the activation-induced flow changes are moderately damped, whereas at high PETCO2, they are nearly lost (deltaCBF (+/-SD): 52% +/- 25%, 68% +/- 22%, 16% +/- 25% at PETCO2 = 20, 40, 70 mm Hg, respectively).

Effect of graded hypo- and hypercapnia on fMRI contrast in visual cortex: quantification of T(*)(2) changes by multiecho EPI.

The sensitivity of functional magnetic resonance imaging (fMRI) in visual cortex to graded hypo- and hypercapnia was quantified in 10 normal subjects using single-shot multiecho echo-planar imaging (Turbo-PEPSI) with eight equidistant echo times (TEs) between 12 and 140 ms. Visual stimulation was combined with controlled hyperventilation and carbon dioxide inhalation to perform fMRI at six levels of end-expiratory pCO(2) (PETCO(2)) between 20 and 70 mm Hg. T(*)(2) in visual cortex during baseline conditions (light off) increased nonlinearly from 20 to 70 mm Hg, from 61.1 +/- 4.2 ms to 72.0 +/- 4.6 ms. Changes in T(*)(2) due to visual stimulation increased 2.1-fold, from 1.2 +/- 0.6 ms at 20 mm Hg to 2.5 +/- 0.7 ms at 50 mm Hg. An almost complete loss of functional contrast was measured at 70 mm Hg. The model of MR signal dephasing by Yablonskiy and Haacke (Mag Reson Med 1994;32:749-763) was used to predict changes in cerebral blood flow (CBF), which were found to be consistent with results from previous positron emission tomography (PET) studies. This study further emphasizes that global CBF changes (due to PETCO(2) changes even in the physiological range) strongly influence fMRI contrast and need to be controlled for.

Modulation of the neuronal circuitry subserving working memory in healthy human subjects by repetitive transcranial magnetic stimulation.

We studied the effect of repetitive transcranial magnetic stimulation (rTMS) on changes in regional cerebral blood flow (rCBF) as revealed by positron emission tomography (PET) while subjects performed a 2-back verbal working memory (WM) task. rTMS to the right or left dorsolateral prefrontal cortex (DLPFC), but not to the midline frontal cortex, significantly worsened performance in the WM task while inducing significant reductions in rCBF at the stimulation site and in distant brain regions. These results for the first time demonstrate the ability of rTMS to produce temporary functional lesions in elements of a neuronal network thus changing its distributed activations and resulting in behavioral consequences.