Ok, so here is why I thought it so bad. They state in the introduction that they wanted to develop this tool for the following reason:
That the ratio of blood ketones to blood glucose will be a more stable biomarker then glucose and ketones separately. So you’d expect in their analysis that they’d actually show this increased value. But they never do. And they have the mouse data, so there is no reason for them not to calculate it. My feeling is that ketone level alone does the trick and makes their GKI irrelevant.
But this is what angered me.
“We present evidence showing that the GKI can predict success for brain cancer management in humans and mice using metabolic therapies that lower blood glucose and elevate blood ketone levels.”
*My emphasis.
They did nothing of the sort and it’s damn irresponsible of them to say that they have.