> We are planing a resting-state fMRI experiment at 3T in which we would like whole-brain coverage with 2.5mm isotropic voxels and use multi-echo EPI to cope with signal loss in regions like the OFC. Obviously this lengthens our TR considerably (~3s). Turning off fat suppression brings our TR back within acceptable limits (2.2 s). I can't think of big problems with this for BOLD imaging; it could increase noise a little bit in areas in which fat is mapped onto brain tissue but there should not be a lot of variance over time I guess. Can anyone comment on whether turning off fat suppression is a good idea for BOLD imaging in general and RS-fMRI in particular?
I think you will just have to try and see what you get. One thing that comes to mind to look out for is movement correction which will not work so well with non-fat-sat images, and which will add to the temporal variance where there is a mixture of fat and water signal. Would using a spatio-spectral pulse be an option?