Story
What your body does in ketosis
When carbohydrate is very low, the liver turns fat breakdown products it cannot burn into ketone bodies. If the body can still release insulin normally, ketones usually stay low; if that insulin brake fails, they climb into acidosis.
Last updated
Overview and key questions
In short: When you eat very little carbohydrate, the liver turns the fat-breakdown products it cannot burn into ketones and sends them into the blood as fuel for other organs.
Educational content, not medical advice. See a clinician about your own case.
Hormones open the gate, fatty acids reach the liver: When insulin in the blood falls and glucagon rises, fat tissue breaks stored fat into fatty acids and releases them into the blood.
Insulin holds the brake: Insulin is the brake on ketone production: it controls both the raw material and the liver's enzyme.
Story path
Chapter 1
The liver changes course
It starts with hormones. Insulin in the blood falls (see How a meal becomes energy), and fat tissue responds by breaking stored fat into fatty acids and releasing them into the blood. The liver takes up those fatty acids and cuts them into small pieces.
These pieces have to pair with a partner molecule to enter the loop in which cells burn fuel. When carbohydrate is scarce, the liver's partner molecule is diverted to making new glucose, so the pieces cannot find a partner. They spill over and are assembled into ketones. Some of the ketones become acetone, which leaves the body in the breath.
When the body can release insulin normally, nutritional ketosis usually stays at fairly low levels. People with diabetes, especially type 1 or those taking an inhibitor, who develop nausea and vomiting, abdominal pain, or deep, fast breathing need to get medical care immediately.
Mechanism · How two hormones open the gate
Insulin presses down in two ways. One is in fat tissue: it makes fat cells break down less fat, so the liver gets no raw material. The other is inside liver cells: it makes them produce less of the key enzyme for making ketones (HMGCS2).
When carbohydrate intake is low, insulin does not rise much after meals, both gates open, and glucagon keeps pushing from the other side. Ketones rising after a full day without food follow this same route; the only difference with a ketogenic diet is that people eat as usual, with so little carbohydrate that insulin likewise stays low.
Mechanism · Why ketones count as overflow
When carbohydrate is scarce, the liver's oxaloacetate has another important job: it is used to make new glucose. With less left for the cycle, more spills over.
The review's authors also state that the amount of acetyl-CoA in the liver does not on its own determine how fast ketones are made, and the mechanisms that fine-tune the rate are not yet fully understood.
The ketones made are mainly acetoacetate and beta-hydroxybutyrate. Acetoacetate is not very stable. It loses a carbon on its own and becomes acetone. Acetone evaporates easily and leaves the body in the breath; when ketone levels are very high, it is the source of the sweet smell on the breath.
Numbers · How much ketone the blood normally holds
Across different nutritional states, ketones supply roughly 5% to 20% of the body's total energy expenditure.
These are typical ranges given in a review, useful for a sense of scale, not cut-offs for judging whether someone is healthy.
Chapter 2
Insulin as the brake
When the brake fails, it is a different story. When a person with type 1 diabetes misses insulin doses or develops an infection, there is not enough insulin; fatty acids pour into the liver, ketones pile up, and the blood turns acidic. This is diabetic ketoacidosis.
When people with diabetes take an inhibitor (a class of glucose-lowering drugs that makes the kidneys pass sugar into the urine), ketoacidosis can occur without high blood glucose. For them, a normal glucose reading is not a sign of safety.
Red flag · When to get care right away
Nausea and vomiting, unable to keep food or drink downAbdominal painDeep, fast breathingA sweet, rotten-apple smell on the breathGrowing drowsiness or confusion, hard to wake
A blood glucose reading that is not high does not rule out ketoacidosis.
If these people want to switch to a low-carbohydrate or ketogenic way of eating, they should talk to a doctor first. A few other groups should ask a doctor first too: people with type 2 diabetes on glucose-lowering drugs such as insulin or sulfonylureas (blood sugar can run low if the dose is not adjusted); people who are trying to conceive, pregnant or breastfeeding; people with chronic kidney disease or severe liver disease; and people with familial hypercholesterolemia (an inherited form of high cholesterol). This is the site's advice, not the finding of any one study.
Clinical · How ketoacidosis is defined
Whether this is ketoacidosis is something a hospital judges from glucose or diabetes history, blood and blood acidity together. Urine ketones can underestimate the severity early on and overestimate it later, so a strip color cannot reassure you.
For comparison, in a small study of 12 healthy adults six weeks into a ketogenic diet, group-average blood beta-hydroxybutyrate moved between about 0.33 and 0.70 mmol/L over the day, with wide differences between people; in ketoacidosis, blood ketones can reach 20 mmol/L. The first figure is only what this small study measured, not a level a ketogenic diet should aim for.
Evidence · Ketoacidosis with normal blood glucose
The authors proposed three mechanisms: the drug clears sugar into the urine without relying on insulin; glucagon rises; and blood volume runs low. This is a case series with no denominator, so it cannot give an incidence and cannot prove cause.
The 2024 consensus gives the bigger picture: about one in ten cases of ketoacidosis are of the normal-glucose type, and most recent ones are linked to SGLT2 inhibitors. In people with type 1 diabetes taking these drugs, about 4% develop it, a 5- to 17-fold higher risk; in type 2 diabetes it is uncommon, about 0.6 to 4.9 cases per 1000 patient-years. The most common triggers are infection and missed insulin.
These numbers belong to people with diabetes only and cannot be applied to people without diabetes who are simply eating a ketogenic diet.
Chapter 3
The first weeks: sodium, water and keto flu
One route involves sodium. When healthy people are given an insulin infusion, less sodium appears in their urine, which shows that insulin makes the kidneys hold on to sodium. Reasoning in reverse, when carbohydrate is low and insulin falls, the kidneys reclaim less sodium, and sodium leaves in the urine taking water with it. This step is an inference; that study only tested insulin going up.
The other route is glycogen: it is stored with water, and when it is used up the water leaves too (see A kilo up or down in a day is not fat).
Online, people describe headache, tiredness, nausea, dizziness and a foggy head and call it keto flu; these self-reports cluster in the first week and become rare after about 4 weeks.
Mechanism · How insulin makes the kidney hold sodium
Glomerular filtration rate, kidney blood flow and blood aldosterone did not change. From this the authors concluded that insulin acts on the diluting segment of the distal nephron and makes it reclaim more sodium.
This was a short experiment in just 6 people with no control group, and it tested insulin going up; more sodium leaving when insulin falls is inferred from it in reverse.
Taking in some sodium and water in the first few days is a practice that starts from this inference; no experiment has said how much. People with high blood pressure, heart or kidney disease, or who take blood pressure medicines or diuretics, should ask a doctor first.
Evidence · Where keto flu reports come from
The most common complaints were a flu-like feeling, headache, tiredness, nausea, dizziness, a foggy head and an upset stomach. Reports clustered in the first week and became rare after about 4 weeks.
These are all self-reports posted online, with no comparison group, and there is no way to confirm that these people were actually in ketosis. A commentary in the same journal argued that its methods do not hold up. It shows that people often mention this feeling; it cannot show how common it is or what causes it.
Chapter 4
Who burns ketones, and what the strip measures
Who burns ketones? Studies of prolonged starvation found that the brain can use ketones as fuel (see Carbs & Fiber). Those measurements come from starvation, not from a ketogenic diet; what the two share is low insulin, with the liver making ketones in both.
Ketone drinks are a different matter. Drinking one does raise blood ketones, but free fatty acids in the blood fall at the same time, which shows that fat tissue is releasing less fat, not more. High blood ketones do not mean you are burning your own fat.
Evidence · What time of day the strip catches it
Blood beta-hydroxybutyrate was lowest at 10 a.m., about 0.33 mmol/L, and highest at 3 a.m., about 0.70 mmol/L.
Urine strips detected more than 90% of the time in the early morning, evening and small hours, but only 58% at 10 a.m. and 50% at 4 p.m. In other words, the same person on the same day may test negative in the morning and positive in the evening.
There were only 12 people, their baseline on their usual high-carbohydrate diet was not measured, and their physical activity varied widely.
Evidence · How much mild ketosis strips miss
Taking blood beta-hydroxybutyrate as the reference, at the three thresholds of ≥ 0.3, ≥ 0.5 and ≥ 1.0 mmol/L, urine strips picked up only 35%, 52% and 76% of cases.
The authors concluded that urine strips are not suitable for detecting mild ketosis, and that blood ketone meters should be used when monitoring is needed. All the participants were postmenopausal women; leaner or younger people, or men, may differ.
People often say the longer they stay on the diet, the paler the strip gets. Why it fades is something we found no reliable study on; what is certain is that the strip measures only acetoacetate, so its color does not stand for blood , and to confirm, look at blood ketones.
Myth · Ketone drinks are not fat burning
Every ketone drink lowered blood glucose, free fatty acids and . The authors attribute this to beta-hydroxybutyrate inhibiting the breakdown of fat in fat tissue. On a ketogenic diet free fatty acids go up; after a ketone drink they go down, the exact opposite direction.
The paper did not measure fat oxidation or body fat, so the point that these drinks are not fat burning is inferred from the suppressed fat breakdown, not measured directly. The studies were small and looked only at a single drink; several authors are linked to a company developing ketone ester products, which the paper discloses.
Some people also say fight aging. One review says only that beta-hydroxybutyrate may be relevant to some diseases and to aging, so the idea that a ketogenic diet or ketone drinks fight aging is not a conclusion that can yet be drawn for people.
Chapter 5
Weight and blood lipids: what the evidence shows
The other thing to know in advance is blood lipids. In many people, (often called bad cholesterol) rises markedly on a ketogenic diet, and how much it rises varies a lot from person to person. One review lists rising LDL among the long-term risks of a ketogenic diet and notes that long-term safety data are still lacking (see Dietary Patterns Compared).
In a trial that 17 people completed, the rise in LDL was visible within 4 weeks. A practice that follows from this: anyone who wants to try the diet can have their blood lipids tested before starting and again a few weeks in. This is an inference, not the paper's advice.
Evidence · The weight gap after a year or more
The ketogenic group lost less than 1 kg more, about 1% of starting body weight. The authors themselves write that in the long term this gap, though statistically significant, is of little clinical significance.
Blood lipids and blood pressure differed too: the ketogenic group's and diastolic blood pressure fell more, and both and rose more.
Adherence in the ketogenic groups of the included trials was low, and by the end of follow-up many people were eating more carbohydrate than the plan allowed; the analysis looked only at blood markers, not at cardiovascular events.
Evidence · How much LDL rises varies by person
The ketogenic diet raised by an average of 1.82 mmol/L and apolipoprotein B by 0.50 g/L; it went up in all 17 women. The sample was small and all female, and 4 weeks is too short to see outcomes such as heart disease.
How much it rises varies a lot from person to person. A 2022 online survey gathered data from 548 people on low-carbohydrate diets: the lower their self-reported body mass index, the more their LDL rose; 100 of them (18%) met all three preset criteria of LDL ≥ 200 mg/dL, ≥ 80 mg/dL and ≤ 70 mg/dL.
This was a voluntary cross-sectional survey whose respondents were leaner and healthier than average, so the proportion cannot be taken as a rate among people on low-carbohydrate diets; letters in the same journal criticized its methods. All it can show is that individual differences are large.
References (19)
- Puchalska, P., & Crawford, P. A. (2017). Multi-dimensional roles of ketone bodies in fuel metabolism, signaling, and therapeutics. Cell Metabolism, 25(2), 262–284. 10.1016/j.cmet.2016.12.022
- Umpierrez, G. E., Davis, G. M., ElSayed, N. A., Fadini, G. P., Galindo, R. J., Hirsch, I. B., Klonoff, D. C., McCoy, R. G., Misra, S., Gabbay, R. A., Bannuru, R. R., & Dhatariya, K. K. (2024). Hyperglycemic crises in adults with diabetes: A consensus report. Diabetes Care, 47(8), 1257–1275. 10.2337/dci24-0032
- Peters, A. L., Buschur, E. O., Buse, J. B., Cohan, P., Diner, J. C., & Hirsch, I. B. (2015). Euglycemic diabetic ketoacidosis: A potential complication of treatment with sodium–glucose cotransporter 2 inhibition. Diabetes Care, 38(9), 1687–1693. 10.2337/dc15-0843
- Yang, M. U., & Van Itallie, T. B. (1976). Composition of weight lost during short-term weight reduction: Metabolic responses of obese subjects to starvation and low-calorie ketogenic and nonketogenic diets. The Journal of Clinical Investigation, 58(3), 722-730. 10.1172/jci108519
- DeFronzo, R. A., Cooke, C. R., Andres, R., Faloona, G. R., & Davis, P. J. (1975). The effect of insulin on renal handling of sodium, potassium, calcium, and phosphate in man. Journal of Clinical Investigation, 55(4), 845–855. 10.1172/JCI107996
- Kreitzman, S. N., Coxon, A. Y., & Szaz, K. F. (1992). Glycogen storage: Illusions of easy weight loss, excessive weight regain, and distortions in estimates of body composition. The American Journal of Clinical Nutrition, 56(1 Suppl), 292S-293S. 10.1093/ajcn/56.1.292s
- Bostock, E. C. S., Kirkby, K. C., Taylor, B. V., & Hawrelak, J. A. (2020). Consumer reports of “keto flu” associated with the ketogenic diet. Frontiers in Nutrition, 7, 20. 10.3389/fnut.2020.00020
- Owen, O. E. (2005). Ketone bodies as a fuel for the brain during starvation. Biochemistry and Molecular Biology Education, 33(4), 246-251. 10.1002/bmb.2005.49403304246
- Cahill, G. F., Jr. (2006). Fuel metabolism in starvation. Annual Review of Nutrition, 26, 1-22. 10.1146/annurev.nutr.26.061505.111258
- Urbain, P., & Bertz, H. (2016). Monitoring for compliance with a ketogenic diet: What is the best time of day to test for urinary ketosis? Nutrition & Metabolism, 13, 77. 10.1186/s12986-016-0136-4
- Gibson, A. A., Eroglu, E. I., Rooney, K., Harper, C., McClintock, S., Franklin, J., Markovic, T. P., Seimon, R. V., & Sainsbury, A. (2020). Urine dipsticks are not accurate for detecting mild ketosis during a severely energy restricted diet. Obesity Science & Practice, 6(5), 544-551. 10.1002/osp4.432
- Stubbs, B. J., Cox, P. J., Evans, R. D., Santer, P., Miller, J. J., Faull, O. K., Magor-Elliott, S., Hiyama, S., Stirling, M., & Clarke, K. (2017). On the metabolism of exogenous ketones in humans. Frontiers in Physiology, 8, 848. 10.3389/fphys.2017.00848
- Hall, K. D., Chen, K. Y., Guo, J., Lam, Y. Y., Leibel, R. L., Mayer, L. E., ... & Ravussin, E. (2016). Energy expenditure and body composition changes after an isocaloric ketogenic diet in overweight and obese men. American Journal of Clinical Nutrition, 104(2), 324-333. 10.3945/ajcn.116.133561
- Hall, K. D., & Guo, J. (2017). Obesity energetics: Body weight regulation and the effects of diet composition. Gastroenterology, 152(7), 1718-1727.e3. 10.1053/j.gastro.2017.01.052
- Hall, K. D., Guo, J., Courville, A. B., Boring, J., Brychta, R., Chen, K. Y., Darcey, V., Forde, C. G., Gharib, A. M., Gallagher, I., Howard, R., Joseph, P. V., Milley, L., Ouwerkerk, R., Raisinger, K., Rozga, I., Schick, A., Stagliano, M., Torres, S., … Chung, S. T. (2021). Effect of a plant-based, low-fat diet versus an animal-based, ketogenic diet on ad libitum energy intake. Nature Medicine, 27(2), 344–353. 10.1038/s41591-020-01209-1
- Bueno, N. B., de Melo, I. S., de Oliveira, S. L., & da Rocha Ataide, T. (2013). Very-low-carbohydrate ketogenic diet v. low-fat diet for long-term weight loss: A meta-analysis of randomised controlled trials. British Journal of Nutrition, 110(7), 1178-1187. 10.1017/S0007114513000548
- Burén, J., Ericsson, M., Damasceno, N. R. T., & Sjödin, A. (2021). A ketogenic low-carbohydrate high-fat diet increases LDL cholesterol in healthy, young, normal-weight women: A randomized controlled feeding trial. Nutrients, 13(3), 814. 10.3390/nu13030814
- Norwitz, N. G., Feldman, D., Soto-Mota, A., Kalayjian, T., & Ludwig, D. S. (2022). Elevated LDL cholesterol with a carbohydrate-restricted diet: Evidence for a "lean mass hyper-responder" phenotype. Current Developments in Nutrition, 6(1), nzab144. 10.1093/cdn/nzab144
- Crosby, L., Davis, B., Joshi, S., Jardine, M., Paul, J., Neola, M., et al. (2021). Ketogenic diets and chronic disease: weighing the benefits against the risks. Frontiers in Nutrition, 8, 702802. 10.3389/fnut.2021.702802
Written from the sources listed above. Checked sentence by sentence against them in . No physician reviews this site: it explains how the body works, and it is not medical advice. How we make content
Spotted an error, or something unclear? Email us. We read every message, check it against the sources, and fix what's wrong.