Fasting

Fasting is going without food long enough to empty glycogen and shift fuel toward tissue protein and free fatty acids. Short gaps between meals are normal. Multi-day fasts and chronic under-eating are stress experiments.

What breaks first

After liver glycogen falls, the body liberates amino acids and free fatty acids. Tryptophan and cysteine rise; thyroid effect falls; cortisol and adrenaline climb. Thymus and lean tissue pay early.

Free fatty acids from PUFA stores block glucose oxidation and amplify stress. See Free fatty acids, Randle cycle, Stress.

A two-day fast impairs oral dextrose tolerance in healthy young men about as badly as a two-day high-fat diet; the same men recover a flat dextrose curve after two days of high carbohydrate.

Practice

Eat enough to keep temperature and pulse up. Use fruit, juice, milk, and protein when stress is high rather than "fasting for autophagy" as a hobby.

A healthy person can go up to about 24 hours on stored glycogen before real deficit starts; thyroid output drops during that stretch specifically to slow tissue breakdown once glycogen runs low.

In one day of fasting, the body stores several ounces of glucose in the form of glycogen. If you're healthy, you can go up to 24 hours before you are short on glycogen. As the body senses stress, lowering the T3 slows the rate of fuel use. When your glycogen is gone, you have to start using your tissue to make glucose out of your muscles.

— Ray Peat

Extended water-only fasting is much more destructive than calorie restriction with minerals. In a 10-day complete water fast, over 80% of the weight lost was muscle and skin; on a 500-600 calorie diet with adequate minerals (sodium, calcium, magnesium, potassium) over the same restriction, nearly all the weight lost was fat.

If you give a little bit of glucose and minerals, sodium, calcium, magnesium, potassium, during the low-calorie period, you can greatly reduce your tissue loss. On a complete water fast for 10 days, they found that 80% or more of the weight loss came from the destruction of muscle and skin. But if they had a 500 or 600 calorie diet with adequate minerals, almost all of the weight loss was in the form of fat loss. So a complete fast is very destructive to the body.

— Ray Peat

Fasting for "detox" specifically backfires: liver glycogen deprivation plus cortisol shuts down the liver's sulfation and glucuronidation systems, the same pathways that clear excess estrogen, so fasting tends to raise circulating estrogen rather than clear toxins. See Estrogen dominance#Opposing estrogen.

If restricting calories for a specific reason, keep some glucose and electrolytes coming in rather than water-only, and prefer restoring thyroid-supportive food (see Hypothyroidism#Practice: diet before supplements) over extending the fast further.

Therapeutic short restriction under medical care is a different claim than lifestyle fasting culture. See Sugar, Thyroid, Intermittent fasting, Harm reduction.

Autophagy and fasting

Fasting increases cellular autophagy during starvation, but the same process occurs during active metabolism. Adequate thyroid function, glucose oxidation, and low PUFA support routine removal of damaged proteins without the cortisol-driven loss of thymus, muscle, and other tissue that accompanies prolonged fasting. Autophagy during starvation is therefore a survival response, not proof that food deprivation improves cellular maintenance.

Calorie restriction

Calorie restriction (CR) lowers food intake chronically, usually 20-40% below normal, on the premise that it extends lifespan. Animal longevity data is mixed once stress costs are counted; in humans, chronic under-eating raises cortisol and lowers thyroid effect.

What calorie restriction actually costs

Clive McKay's foundational 1930s life-extension experiments predate most work on which specific nutrients, rather than calories generally, drive the effect:

Although Clive McKay's studies of life extension through caloric restriction were done in the 1930s, only a few studies have been done to find out which nutrients' restriction contributes most to extending the life span.

— Ray Peat

Under-eating does not shrink the body evenly, stress, starvation, and even hibernation all burn lean tissue faster than fat:

The main bulk of an animal's body consists of water, protein, fat and bones. Fat tissue and bone are metabolically more quiescent than the protein-water systems. During stress or starvation, or even hibernation, animals lose lean mass faster than fat.

— Ray Peat

Chronic calorie restriction as a lifestyle, rather than a short deliberate deficit, risks losing muscle and organ tissue rather than fat specifically, on top of the cortisol/thyroid tradeoffs.

Practice (Calorie restriction)

Prefer raising metabolic rate and food quality over long semi-starvation. See Fasting, Cortisol, Thyroid.

Intermittent fasting

Intermittent fasting and other long gaps between meals empty liver glycogen (stored sugar). Free fatty acids rise. Active thyroid hormone T3 falls. After glycogen is gone, the body burns protein. Thymus first, then muscle and liver.

After the liver's glycogen is depleted, fasting destroys the tissues, starting with the thymus, then the muscles and liver.

— Ray Peat

Glycogen window

A healthy liver holds enough glycogen for a limited gap. Past that, muscle protein becomes sugar. T3 falls and slows burn rate. That slows self-digestion. It is metabolic depression.

In one day of fasting, the body stores several ounces of glucose in the form of glycogen. If you're healthy, you can go up to 24 hours before you are short on glycogen. As the body senses stress, lowering the T3 slows the rate of fuel use. When your glycogen is gone, you have to start using your tissue to make glucose out of your muscles. If your thyroid didn't slow down during this prolonged stress, you would eat your whole body up in just a few days. But since the depressive process turns off metabolism, it slows the rate of degeneration.

— Ray Peat

A healthy person can go 12 to 18 hours. High body PUFA turns a sugar dip into circulating polyunsaturated free fatty acids that damage tissue.

A healthy liver can store glycogen, so a healthy person can go 12 to 18 hours without eating. But if you have too much PUFA in your body, as soon as blood sugar dips, you pull PUFA into circulation and start damaging tissues.

— Ray Peat

Coffee, aspirin, and solid thyroid function make a meal gap less costly.

Largest human trial: muscle loss, not fat-specific loss

The TREAT trial (UCSF, ~116 participants, one of the largest randomized human trials of time-restricted eating to date) found 16:8 time-restricted eating produced no significant weight loss advantage over three meals a day (-0.94 kg TRE vs -0.68 kg control, between-group difference not significant), while the fasting group showed a significantly larger drop in appendicular lean mass index than controls. The trial's lead investigator, a cardiologist who had personally practiced time-restricted eating for years, stopped the practice after seeing the results.

Cochrane review: no meaningful advantage over regular dieting

A 2026 Cochrane systematic review pooled 22 randomized controlled trials (1,995 participants) comparing intermittent fasting against regular dietary advice for adults with overweight or obesity. Compared to regular dietary advice, intermittent fasting showed little to no difference in weight loss, in reaching a 5% body weight reduction, or in quality of life. None of the differences reached clinical significance. Intermittent fasting is not a superior weight-loss method in the TREAT trial, and it carries an added tissue-loss cost.

Night and diabetes

Every night blood sugar falls. Free fatty acids rise. Fat burning wastes oxygen. Lactate rises. Protein can become sugar. Mitochondria take damage. Bone mineral appears in morning urine. Longer voluntary fasts deepen that night pattern.

Blood sugar normally falls during the hours of darkness. This is compensated by a rise in free fatty acids in the bloodstream. That is the same pattern that happens when you are under stress. When glycogen is depleted, you mobilize free fatty acids to oxidize for energy. This requires more oxygen and is inefficient. … The longer you go without eating, the more you rely on inefficient fat oxidation.

— Ray Peat

Fasting lowers blood sugar. That drop is stress. Lasting diabetes repair means lower polyunsaturated fat, butter or coconut for cooking, and steady carbohydrate so sugar can support pancreas repair.

It will make your blood sugar go down, but anything that you can call fasting is going to be putting you under stress. The long-range thing that corrects chronic diabetes is to get your polyunsaturated fats (PUFA) lower, to completely eliminate them from your diet as far as possible.

— Ray Peat

Ketogenic fat-burning states lock free-fatty-acid block of full sugar oxidation to CO₂. See Randle cycle and Diabetes.

Eat on a regular schedule. Keep glycogen and blood sugar steady. If a long gap is forced, take a little carbohydrate and minerals (sodium, calcium, magnesium, potassium). Lower body PUFA. Night dips cost less. Judge by warmth, pulse, sleep, mood, and thyroid signs.

Refeeding

Refeeding syndrome is a set of dangerous fluid and electrolyte shifts, with phosphate, potassium and magnesium falling sharply, when nutrition returns after severe starvation. Hospital protocols exist for a reason, and refeeding after extreme restriction should happen under medical care. Ordinary diet repair after mild dieting still wants salt, sugar and minerals, without the same crash risk.

Timing and definition

Refeeding syndrome usually appears in the first 72 hours after nutrition restarts. A systematic review of 45 studies covering 6,608 patients found that its definition varies from study to study, most relying on blood electrolyte changes alone, and reported incidence ranged from 0% to 80% depending on the definition and the patients studied. Risk factors largely matched the UK NICE guideline, with older age and tube feeding as additional ones, but evidence for firm links to adverse outcomes and for prevention protocols was weak.

How fast to refeed

The trials do not uniformly support the old "start low and go slow" rule, but only in certain groups. In a 13-ICU randomised trial of 339 critically ill adults who developed refeeding syndrome, protocolised calorie restriction did not significantly change days alive after ICU discharge (44.8 versus 39.9 days,, yet more patients were alive at day 60 (91% versus 78%,. In adolescents with anorexia nervosa the opposite pattern showed up: a systematic review of 20 studies and 2,191 patients found a true clinical case of refeeding syndrome in only one study, and no evidence that a high-calorie start raised the risk. Low body mass index at admission predicted low phosphate better than calorie intake did.

Practice (Refeeding syndrome)

After a prolonged fast or severe restriction, get bloodwork before and during the first days of eating, and do not self-manage a large jump in food. Blood phosphate, potassium and magnesium are the values to watch, and supplementation is guided by those results. See Fasting and Anorexia.