The first functional MR images were collected by Belliveau and colleagues at the Massachusetts General Hospital (MGH) around 1989 and 1990 and published in 1991 . The technique these investigators chose to use was one that had also been pioneered at the MGH (no accident)  using intravenous administration of a bolus of paramagnetically labeled salt, Gd(DTPA)2". This technique allows the measurement of relative cerebral blood volume based upon the fact that the contrast agent remains in the intravascular space. Since the contrast agent is chosen to have a high magnetic susceptibility a large gradient is induced between the blood and the brain tissue. These gradients, which spread out far beyond the actual spatial volume distribution of the blood vessels, decrease in turn the relaxation rates of the water spins that diffuse through the gradients thereby affecting a relatively large proportion of the total water signal (i.e., in excess of the actual CBV). This renders the technique more sensitive than it might otherwise be, but at a cost of only being able to measure the relative, rather than absolute, cerebral blood volume. At commonly used concentrations of agents and imaging parameters, the total signal loss observed as the bolus passes through the brain can be close to 100%. This signal rapidly returns to baseline. Then, standard tracer kinetic analyses can be applied to the signal to derive parameters related to cerebral hemodynamics. The area under the curve is roughly proportional to CBV, while the peak drop is roughly proportional to CBF.
The main limitations of this technique are twofold. First, there is the requirement of injection of at least two doses of contrast agent (before and after the stimulus). Even though Gd(DTPA)2" is extremely nontoxic there is still a minimal risk associated with its use. Second, in order to collect multiple time points one must perform multiple injections. This quickly becomes prohibitive, and dramatically limits the statistical power of the technique. Since this technique is no longer used for functional imaging-type experiments we will not consider it further. Interested readers can obtain further details from some of the original reports [1,64]. It should be pointed out, however, that this technique is still widely used for measuring hemodynamic parameters in a clinical setting for many types of cerebral pathologies, such as strokes and tumors. Its direct descendent is, however, being used more frequently, as described below.
Was this article helpful?