What "ketogenic" actually means
A ketogenic diet is not the same as a low-carbohydrate diet. In the classical definition used for childhood epilepsy, 80–90% of calories come from fat, protein is held near 1 g per kilogram of body weight, and carbohydrate is limited to about 10–15 g a day. The goal is measurable ketosis, not simply fewer carbs.
Adult cats have a published ceiling for fat. The 2006 NRC nutrient requirements list a safe upper limit of 82.5 g of fat per 1,000 kcal, which works out to roughly 74% of calories. A classical ketogenic formula is designed to sit above that line. That single comparison sets the tone: a diet used medically in children is one the feline reference tables do not endorse.
Cats do not fast the way people do
Humans and dogs shift into ketosis within a day or two of carbohydrate restriction because fat is their main fuel. Cats are built differently. They are obligate carnivores with high baseline gluconeogenesis: much of their blood glucose is manufactured from amino acids, and their protein requirement is correspondingly high. The NRC recommended allowance for adult maintenance is about 50 g of protein per 1,000 kcal, and complete foods usually deliver 26% or more protein on a dry matter basis.
When a cat stops eating, the liver mobilises fat faster than it can export it. Feline hepatic lipidosis can develop after roughly 2–7 days of anorexia, and overweight middle-aged cats are the classic victims. A diet built on near-zero carbohydrate and a very high fat load is being tested on an animal whose fasting metabolism is already fragile.
What cat studies have actually measured
Two feline datasets are usually cited for ketones; neither is a ketogenic diet trial. Trevizan and colleagues fed 19 healthy adult cats a coconut-oil diet, a medium-chain triglyceride source, for 9 weeks; blood beta-hydroxybutyrate and plasma triglycerides rose, and palatability did not suffer. Jackson and Jewell randomised 64 healthy cats to four diets for 28 days, one of which replaced pork fat with 7% medium-chain triglycerides on a dry matter basis; ketone production rose, most clearly when MCT was combined with fish oil.
Both show that cats can produce ketones. Neither tested a disease outcome or used a formula with 80–90% of calories from fat. A 2024 systematic review of ketogenic diets in dogs and cats screened PubMed, Embase and Scopus and found seven canine papers plus three conference abstracts; its authors concluded that no clinical study of ketogenic diets in cats existed, and no peer-reviewed placebo-controlled cancer study in either species.
The diet that did produce remission
The strongest feline evidence points somewhere else entirely. A 2026 two-centre randomised trial enrolled 72 overweight, insulin-treated diabetic cats and fed every one of them a therapeutic diabetic diet for 12 weeks; half were also calorie-restricted toward about 2% weight loss per week. Remission was reached by 16 of 32 restricted cats (50%) versus 12 of 40 controls (30%), a 2.1-fold difference with P = .04. Restricted cats lost 7.2% of body weight against 2.7%, their glucose variability fell 45%, and their insulin dose dropped 36% while controls needed 28% more.
Look at what those cats ate: the dry diet was 49.4% protein, 24% fat and 26.6% nitrogen-free extract as metabolisable energy; the wet diet was 63.2% protein, 25.3% fat and 11.5% carbohydrate. That is a high-protein, moderate-fat, low-carbohydrate diet, not a ketogenic one. The 2025 iCatCare consensus puts the same target in numbers: under 25% carbohydrate on a dry matter basis, under 15% of metabolisable energy, or under 5 g per 100 kcal. A 2022 JAVMA review of feline carbohydrate research settled on a broader 5–26% ME range, and noted one trial in which 3.5 g carbohydrate per 100 kcal took 68% of cats off insulin versus 41% at 7.6 g.
Where 80% fat goes wrong in a cat
Liver. Any diet change that a cat refuses can trigger hepatic lipidosis within days. A high-fat formula is also the least palatable option for many cats, turning a fussy week into a medical admission.
Pancreas. In dogs, dietary fat is a recognised risk factor for pancreatitis. In cats the ACVIM 2021 consensus describes a usually idiopathic disease associated with diabetes, chronic enteropathy, cholangitis and hepatic lipidosis, not with a measured fat intake. That is not a licence to assume high fat is harmless, nor to blame it for every flare.
Nutrient space. Fat supplies 9 kcal per gram against 4 for protein. At 80–90% of calories from fat there is very little room left for protein, taurine, B vitamins and minerals, which is why human ketogenic diets need supplementation and why growing kittens, pregnant queens and cats with kidney or liver disease are the worst candidates.
Four diet profiles side by side
| Diet profile | Fat (% ME) | Protein (% ME) | Carbohydrate | Feline evidence |
|---|---|---|---|---|
| Classical ketogenic | 80–90 | about 10–15 | under 5% ME | No published cat trial |
| NRC adult safe upper limit | about 74 | — | — | Reference ceiling, not a target |
| Wet diabetic diet, 2026 trial | 25.3 | 63.2 | 11.5% ME | 72 cats, 50% vs 30% remission |
| iCatCare 2025 target | not specified | high | under 15% ME, under 5 g/100 kcal | Consensus guideline |
If you want fewer carbohydrates, do this
- Choose a complete wet food with under 10% metabolisable energy from carbohydrate, and check the figure in g per 100 kcal rather than trusting a front-of-pack claim.
- Transition over 7–10 days in roughly 25% steps; a sudden swap is the commonest cause of refusal.
- Weigh weekly on a 5 g scale and aim for 0.5–2% body weight loss per week. Federal survey data put 61% of American cats in the overweight or obese range, so weight is usually a bigger lever than macronutrient ratio.
- Never fast a cat to induce ketosis. Two days without food is a liver risk, not a therapeutic step.
- Measure blood beta-hydroxybutyrate only when a veterinarian asks for it. Healthy cats usually sit under 0.2 mmol/L, fasted or ketone-fed animals can reach about 0.9, and the 2026 AAHA diabetes guidelines treat a reading above 2.4 mmol/L in a cat on an SGLT2 inhibitor as a reason to switch back to insulin.
The short answer
Ketogenic diets are a legitimate area of veterinary research, but in cats the evidence is a handful of medium-chain triglyceride feeding studies and zero disease trials. If your goal is diabetes control, the tested route is a high-protein, low-carbohydrate wet diet with controlled calories, insulin adjustment and home glucose monitoring — the protocol behind that 50% remission rate. If your goal is cancer control, no feline study supports it.
This article is for information and does not replace veterinary diagnosis or a nutrition plan built for your cat. Ask your veterinarian before changing diets, especially if your cat has diabetes, kidney disease, liver disease or is pregnant.
Terms mentioned in this article
Frequently asked
Can a ketogenic diet cure my cat's cancer?
No feline study supports that claim. A 2024 systematic review found seven dog papers and three conference abstracts on ketogenic diets, and no clinical study in cats; no peer-reviewed placebo-controlled cancer trial exists in dogs or cats either. The Warburg effect, the rationale for carbohydrate restriction in human oncology, has not been confirmed in either species. Weight loss and malnutrition during cancer treatment are real risks, so discuss any diet change with the treating oncologist rather than starting one alone.
Is a high-fat diet safe for a diabetic cat?
The 2026 randomised trial that produced a 50% remission rate used 24% fat and 49.4% protein as metabolisable energy in dry food, and 25.3% fat with 63.2% protein in wet food. No trial has tested an 80-90% fat formula in diabetic cats. The 2025 iCatCare consensus asks for under 25% carbohydrate on a dry matter basis, under 15% metabolisable energy, or under 5 g per 100 kcal, combined with insulin adjustment and home glucose monitoring.
How would I know whether my cat is in ketosis?
Blood beta-hydroxybutyrate, not urine dipstick, is the practical measure. Healthy cats usually sit below 0.2 mmol/L, and fasted or ketone-fed animals can reach about 0.9 mmol/L. The 2026 AAHA diabetes guidelines treat a reading above 2.4 mmol/L in a cat receiving an SGLT2 inhibitor as a reason to stop that drug and switch to insulin. Ketosis is not a target to chase at home; a high reading in a diabetic cat is a reason to call the clinic.