How to break a long fast

The first meal after twenty-four hours matters more than people think. A walk through what your gut and pancreas have been doing while you weren't eating, and what they need from you on the way back.

In brief: after a fast of 24 hours or more, break it with a small, gentle, low-sugar first meal — bone broth, eggs, cooked vegetables, protein and fat rather than refined carbohydrate. Stop before you are full, then wait one to two hours before eating more. A large or sugary first meal is what causes most post-fast nausea, bloating, and energy crashes.

Fasting and refeeding aren't symmetrical. Going twenty-four, thirty-six, seventy-two hours without food asks the body to switch fuels, slow the gut, and quiet the pancreas. Coming back asks for all of those to wake up — gracefully, in the right order. A first meal that's too big or too sweet skips that wake-up and lands a shock the body can react badly to.

Most I felt sick after my long fast stories aren't about the fast. They're about lunch.

What the gut has been doing

Without food, the small intestine doesn't shut down — it stays awake and runs the migrating motor complex, the cleaning waves that sweep residue from one meal to the next. But it does downshift. Bile production drops. Stomach acid output drops. Pancreatic enzymes — lipase, proteases, amylase — sit at lower baseline production. The cells lining the gut even recede slightly; absorptive surface area decreases.

None of this is damage. It's adaptation. But it does mean the digestive machinery you're about to ask for help is running quietly when you walk into the kitchen, and a large, complex meal will land before it's fully spun up.

The result is usually one of three things: bloating and gas (overload on a sluggish gut), nausea (rapid distension), or a delayed, drowsy crash an hour or two later — the tail of an outsized glucose swing, which the next section takes apart.

What insulin does on the comeback

After a long fast, baseline insulin is at its lowest in weeks and the pancreas hasn't had to release much. What happens when the first meal arrives is not quite the story usually told. Measured directly, the early insulin response comes back slower rather than harder: in healthy, non-obese people given a glucose load after thirty-six hours of fasting, insulin at the thirty-minute mark was lower than it had been after an ordinary overnight fast, while blood glucose two hours in was substantially higher — and this in people whose insulin sensitivity had improved on every index the study used. So a carbohydrate load after a long fast produces a bigger and longer glucose swing than the same load would have produced before it, and the shaky, sweaty, sleepy hour people call reactive hypoglycemia sits at the far end of that swing.

This is uncomfortable but not dangerous in healthy people. It does, however, undo some of what the fast was teaching the body. The point of pushing into a fasted state is usually to teach the system to handle fuels gently. A 60-gram-carb refeed teaches it the opposite.

Refeeding syndrome — a real but distant risk

The most serious version of this story is refeeding syndrome: when severely depleted patients (typically after more than five days of starvation, or in chronic malnutrition) are given large carbohydrate meals, the insulin surge drives phosphate, potassium, and magnesium rapidly into cells. Serum levels plummet, and arrhythmias, seizures, even death can follow. Hospital protocols for refeeding malnourished patients are deliberately slow because of this — historically, it's why post-war and post-famine relief required medical supervision.

There is a fourth substance in that story, and it isn't an electrolyte. Thiamine — vitamin B1 — is a cofactor the body spends in order to metabolise glucose, so the moment carbohydrate arrives after a long gap, demand for it climbs in a body whose stores may already be thin. That is why clinical protocols give thiamine before the first calories rather than alongside them: the shortfall is created by the feeding, not by the fast. It also announces itself differently from the electrolyte shifts. Low phosphate and potassium show up in the heart and the muscles; thiamine deficiency shows up in the nervous system, as the confusion, eye-movement changes and unsteadiness of Wernicke's encephalopathy.

For a healthy adult breaking a 24- to 72-hour fast, refeeding syndrome itself is not a meaningful risk. But the physiology is the same in miniature, which is why the practical advice — small, simple, slow — has the same shape regardless of scale.

What a good first meal looks like

The pattern that tends to work, across both clinical practice and informal fasting traditions:

• Small. A quarter to a third of a normal meal portion. The goal is to wake the system, not to feed it.
• Cooked, gentle. Eggs, bone broth, well-cooked vegetables, soft fish, avocado. Skip raw fibrous vegetables and dense cuts of meat in the first sitting — they're slow even when the gut is sharp.
• Low in sugar and refined carbs. Not zero carbs, but not the bowl of pasta. A small portion of complex carbohydrate (oats, sweet potato, rice) is fine. Sweets and fruit juice are the worst single mistake.
• Protein-and-fat-leaning. Both digest more slowly than carbohydrate, which blunts the insulin response and gives the pancreas a gentler curve. The longstanding break a fast with bone broth advice is right mostly because it does this — protein, fat, salt, warmth, no sugar.

A real example: a small bowl of bone broth, then half an avocado with two scrambled eggs and a pinch of salt. Twenty minutes of eating. Then wait.

The wait

The single most useful rule about refeeding is to stop eating before you're full on the first sitting. Then wait one to two hours before the next meal. The wait gives the gut time to register what it just received and ramp enzyme production. The second meal can be larger and more normal. By the third meal, you're usually back to your regular eating pattern without consequence.

The temptation after a long fast is to eat everything you've been thinking about for two days. Don't. The food isn't going anywhere, and the way you come out of a fast affects how you feel for the next twelve to twenty-four hours.

What to avoid in the first six hours

A short list of things that reliably make the comeback unpleasant:

• A very large meal of any kind
• Sugar (juice, sweets, fruit on an empty stomach in any quantity)
• Alcohol — the liver is already busy reorienting; alcohol on top is rough
• A double espresso with nothing else — caffeine on a fully empty gut is harsher than usual
• Spicy food — the gut lining is more reactive than after a normal meal

After six to twelve hours of measured refeeding, normal eating resumes without incident.

The longer the fast, the slower the return

A 16-hour fast can be broken with a normal meal; the body barely notices. A 24-hour fast benefits from a small first meal and waiting an hour before more. A 36- to 48-hour fast wants two or three small meals over the first day. Anything past three days starts to cross into territory where slower is genuinely safer, and a clinician's eye is useful — not as a formality, but because the rules begin to matter.

A clean, careful refeed is part of the fast. Skipping it is what makes long fasts feel reckless, even when they aren't.

Five days and beyond

Past five days the published guidance stops being general and becomes specific: anyone who has eaten little or nothing for more than five days should be refed at no more than half of normal requirements for the first two days. Go slower than feels necessary, stage it, and have a clinician who can see you.

That line comes from Mehanna's BMJ review, and its wording is worth keeping intact. Requirements means a day's total intake, not the size of a plate, and the instruction was written for people being fed in a hospital. It is quoted here as the shape of the caution, not as something to measure out in a kitchen.

The largest published series of long water-only fasts makes the same point from the other side. Finnell and colleagues reviewed 768 medically supervised water-only fasting visits, median length seven days. Refeeding was not an afterthought in that programme: it ran for half the length of the fast, staged across five phases, beginning with juice and moving gradually onto solid plant foods. Under that protocol the refeed was the calmer half — the average severity of adverse events during refeeding came out measurably below the average during the fast itself.

That result is the argument for a staged return, not a reason to relax about one. The refeed half was uneventful because it was protocolised, supervised, and given as much time as it was. At this length the fast is the easy half to plan; the return is the half worth planning harder.

When to stop and get help

New confusion, palpitations or an irregular heartbeat, unusual muscle weakness, swelling, or breathlessness in the days after you start eating again — stop and get medical help.

Those are the signs clinical teams watch for in the first five days of feeding, which is the window in which refeeding complications appear — in one series, low phosphate turned up an average of 1.9 days after feeding restarted. No single one of them is a diagnosis, and this page cannot tell you what any of them means for you. That is the whole point of the list: it names the moment to stop reading and speak to someone who can examine you.

Sources

1. Mehanna HM, Moledina J, Travis J. "Refeeding syndrome: what it is, and how to prevent and treat it." BMJ, 2008. doi:10.1136/bmj.a301
2. Friedli N, Stanga Z, Sobotka L, et al. "Revisiting the refeeding syndrome: results of a systematic review." Nutrition, 2017. doi:10.1016/j.nut.2017.01.011
3. Crook MA, Hally V, Panteli JV. "The importance of the refeeding syndrome." Nutrition, 2001. doi:10.1016/S0899-9007(01)00542-1
4. Cahill GF Jr. "Fuel metabolism in starvation." Annu Rev Nutr, 2006. doi:10.1146/annurev.nutr.26.061505.111258
5. Persaud-Sharma D, Saha S, Trippensee AW. "Refeeding Syndrome." StatPearls, 2026. NBK564513
6. Finnell JS, Saul BC, Goldhamer AC, Myers TR. "Is fasting safe? A chart review of adverse events during medically supervised, water-only fasting." BMC Complement Altern Med, 2018. doi:10.1186/s12906-018-2136-6
7. Tripolt NJ, Hofer SJ, Pferschy PN, et al. "Glucose metabolism and metabolomic changes in response to prolonged fasting in individuals with obesity, type 2 diabetes and non-obese people — a cohort trial." Nutrients, 2023. doi:10.3390/nu15030511
8. Related reading: the electrolytes guide and intermittent fasting schedules.