The red blood cells (erythrocytes, red corpuscles) have no mitochondria and so depend on a supply of glucose and I understand there are other specific cells with a similar need for glucose.
Prof. Bikman has challenged the notion that the brain needs glucose at all, saying that he’s never seen a paper demonstrating this. On the other hand Cahill claims this as a result of his research, and you’d expect Bikman to have read all of Cahill’s work. Dr. Ede says that there are certain parts of neurons that need glucose, because they are too narrow to contain mitochondria, but gives no reference in the video I saw.
And just to confuse things, there is a paper (which I can’t find a link to at the moment, I believe the first author was Drenick) reporting an experiment in which they produced extremely low serum glucose values in fasting subjest (hence in ketosis) with no ill effect.
So my conclusion from all this would be that the brain may very well need a small amount of glucose, but not as much as we’ve been led to believe necessary. The figure of 30% of brain energy needing to come from glucose is certainly too high.
ETA: Hyper is Greek for “above,” hypo for “below.” The Latin equivalents are super and sub.