An attractive feature of measurement of rCBV, as opposed to BOLD, is that rCBV changes should be independent of field strength, whereas BOLD signal changes are extremely field strength dependent. This means that drug-induced changes observed at different fields can be compared for IRON, but not for BOLD. We investigated whether or not such a contention was true, by comparing the activation due to two drugs targeting the dopamine transporter: cocaine and methylphenidate. We then compared the measured rCBV changes at 2 T, 4.7 T, and 9.4 T. As expected, the interanimal variability was much larger than the variation in rCBV as a function of field strength . Both experimental data and simulations showed that the CNR advantages of IRON persist even at field strengths as high as 4.7 T by a factor of 3 [87,94]. At very high fields such as 9.4 T, BOLD CNR increases to the point where it is competitive with IRON. However, there are some additional considerations that render the IRON technique advantageous even at 9.4 T. For instance, echo times and iron oxide concentrations can be traded off. Long echo times can be traded off for less iron, conversely if one desires short echo times, then more iron can be used. The optimal CNR for BOLD comes when TE ~ T2, however, at 9.4 T T2 in the brain is quite short being approximately 20 msec. Gradient-echo images collected with a T2 of 20 msec at 9.4 T will be of poor quality in many regions including those near sinuses. By using a higher concentration of iron oxide contrast agent, one can use ultrashort echo times and maintain constant CNR, at levels still higher than BOLD.
Since MRI in general, and functional or pharmacological MRI specifically, is almost always in need of more SNR as well as CNR (to push spatial and temporal resolution), we believe that the huge CNR increase of IRON over BOLD at most field strengths means that IRON is the technique of choice in animal models. In fact, at field strengths lower than 7 T one could even argue that the use of BOLD causes far more animals to be used than is necessary from a scientific and ethical point of view. Not to put a fine point on it, but at 4.7 T, BOLD imaging would require averaging data from about six animals to attain the same CNR as with a single drug challenge experiment using IRON. Thus, one could argue from the point of view of using the fewest animals with the least amount of suffering that BOLD experiments are unethical when the alternative to use IRON exists!
Was this article helpful?