We now know that the oft-repeated “your brain only runs on glucose!” is wrong . I’ve mentioned it before , and anyone who’s taken the time to get fat-adapted  on a low-carb Primal eating plan  intuitively knows that your brain doesn’t need piles of glucose to work, because, well, they’re using their brain to read this sentence. Obviously, you eventually adapt and find you have sufficient (if not much improved) cognition without all those carbs. That said, some glucose  is required, and that’s where people get tripped up. “Glucose is required” sounds an awful lot like “your brain only uses glucose” which usually leads to “you need lots of carbs to provide that glucose.” And that’s the question today’s edition of “Dear Mark ” finds itself attempting to answer: how much glucose is required?
Let’s get to it.
I have a little problem. Even though I’m able to function at work, maintain conversations, and go about my daily life without having segments of my brain suddenly stop working while eating Primal, my friends are worried about my brain. All they know is that the brain needs glucose. What can I tell them? How much glucose does my brain actually require to keep working?
I wouldn’t be too hard on your friends. They mean well and it’s a common misconception. Instead of chiding them, rubbing their faces in the knowledge that you can function quite adequately on a high-fat diet , educate them.
How much glucose the brain requires depends on the context. There’s not one single answer.
If you’re on a very high fat, very low carb diet – like a traditional Inuit diet – your brain will eventually be able to use fat-derived ketones  for about 50-75% of its energy requirements . Most ketones are produced in the liver, but astrocytes in the brain also generate ketones themselves for use by neurons . You think we’d have that kind of set up in our brains if ketones weren’t useful to have around? If all we could do was burn glucose up there, what would be the point of even having localized ketone factories? Anyway, since the brain can use about 120 grams of glucose a day (PDF ), that means you’d still need at least 30 grams of glucose while running on max ketones.
If you’re merely on a lower carb diet – staying under 150 grams per day or so – or eating medium chain triglycerides (coconut oil , MCT oil) to directly generate ketones, you’ll have access to ketones without being in full-blown ketosis, and your brain will be accessing some of them for energy. Take the story of Dr. Mary Newport , who lessened her husband’s Alzheimer’s symptoms simply by adding a couple tablespoons of coconut oil  to his regular diet. The MCTs in the coconut oil were converted to ketones, which his brain began using. You’ll probably need more than 30 grams of glucose, but you won’t need the full 120 grams on a lower carb Primal way of eating (especially if you eat some coconut).
If you’re involved in strenuous exercise, your brain will be running primarily on lactate. Yep, lactate – that unwanted metabolic byproduct of muscle metabolism. During exercise, when the muscles are using up most of the available glucose to lift things and move a bunch of intelligent primate flesh through three dimensional space, and where inadequate oxygen (hence breathing hard) leads to incomplete glucose and pyruvate breakdown and increased lactate levels, the brain will draw upon lactate  as a direct energy source. Not only that, but lactate appeared to make the brain run more efficiently, more snappily, and when both are available, the brain prefers lactate over glucose . Other research  has found that the brain also prefers lactate in the hours and days immediately following a traumatic brain injury. I’m not sure how much glucose the brain requires when it’s accessing lactate, but it’s definitely fewer than 120 grams.
Of course, even when you need some glucose, that glucose needn’t necessarily come from dietary carbohydrate. It can famously come from gluconeogenesis, the process by which the liver converts amino acids into glucose. It can also come from glycerol, a byproduct of fat metabolism. In deep fasting  situations, glycerol can contribute up to 21.6% of glucose production , with the rest presumably coming from gluconeogenesis. The glycerol can come from both dietary fat and adipose tissue (the authors of that glycerol fasting study even suggest that fasting burns body fat in order to provide glycerol for glucose production), while the amino acids can come from dietary protein (if you’re eating ) or muscle (if you’re starving).
Overall, recent research into the metabolic demands of brain slices  (“living” pieces of brains isolated and used for research) shows that incorporating other energy substrates – ketones, lactate, or even pyruvate – into the glucose solution improves oxidative metabolism and neuronal efficiency. Before you say “but this was in vitro , my brain’s not sliced up and submerged in a weird syrupy solution,” know that the whole point of the study was to better replicate the conditions of the kind of real, actual, living, thinking brains we find in human heads. The authors note that the glucose-only solution normally used to fuel brain slices in other studies is limited, because “in the intact brain, complex machinery exists that coordinates energy substrates delivery and adjusts energy substrate pool composition to the needs of neuronal energy metabolism.” In other words, glucose solution is an easy, dependable way to fuel brain slices, but it’s an incomplete representation of how brains work in heads. The authors conclude that “in slices as well as in vivo, the ability of glucose to maintain energy metabolism is limited and neuronal energy supply should be supported by other oxidative substrates.”
So, a healthy, efficient brain is one that draws on several different fuels. A healthy, efficient brain is one that uses ketones (and perhaps lactate and other fuels) to spare some glucose. A complete reliance on glucose indicates an underachieving brain, a brain that could do so much better, a brain that could really use a coconut milk curry and some intense exercise  every now and again. As far as we can tell, then, the absolute physiological minimum is 30 grams of glucose. I wish I could provide hard numbers for some of the other contexts beyond near carnivory  (like basic 150 grams carbs Primal eating with coconut  or maybe figuring out how to rely on lactate fueling), but the numbers don’t really matter in practice. What matters is that our brains don’t need the full 120 grams of glucose, especially if we’re following a Primal Blueprint eating plan.
I hope that helps.
Questions? Comments? Concerns? Leave them here. Thanks for reading!