Ghrelin rises, leptin won't hold still, and cortisol is the one that reliably moves#
Ask what a short night does to your hunger hormones and the popular answer is a single tidy sentence: sleep loss drops leptin (the hormone that signals fullness), raises ghrelin (the one that signals hunger), and between them they make you overeat. That sentence traces to a real, carefully run experiment — but it has been sanded into a certainty its own evidence never carried. Across the wider literature ghrelin does tend to rise; leptin moves in whichever direction the study design pushes it, sometimes falling and sometimes climbing; cortisol goes up the evening after a lost night more dependably than either; and when you pool the controlled trials, the hormone signal shrinks toward nothing even as the extra eating shows up regardless.
So "controls" is the wrong verb, and that is the correction this page exists to make. Sleep nudges a set of appetite hormones, unevenly and sometimes contradictorily, and those hormones are better understood as a correlated signal than as the lever that drives the behaviour. The behaviour itself — the late grazing, the pull toward calorie-dense food, the few hundred extra calories — is real and well measured, and each leg of it has its own article: why you crave junk food when you're tired, how much more you eat the next day, and how that surplus slowly becomes weight. What follows is the chemistry underneath all three, audited one hormone at a time — because the parts that hold and the parts that don't are not the parts most people expect.
Ghrelin: the signal that mostly behaves#
Ghrelin is the clearest case, and even it is not airtight. The origin experiment put 12 healthy young men through two nights of 4 hours in bed and two of sleep extension, in a randomized crossover with diet and activity held constant; the short nights raised ghrelin 28% and cut leptin 18%, alongside a 24% rise in hunger and a 23% rise in appetite — sharpest for calorie-dense, high-carbohydrate foods, where the wanting jumped 33% to 45%1. Twelve men, two nights, a severe dose: a mechanism shown cleanly, not a population effect you can read off your own week.
A single night can move it. Nine normal-weight men kept fully awake for one night showed morning ghrelin 22% higher than after seven hours of sleep (0.85 versus 0.72 ng/mL; P = 0.048), with markedly stronger hunger — and, tellingly, no change in leptin at all3. The direction also tracks habitual sleep across a whole population: in 1,024 adults from the Wisconsin Sleep Cohort, sleeping 5 rather than 8 hours predicted 14.9% higher ghrelin and 15.5% lower leptin, with body-mass index lowest at 7.7 hours2. That last study is cross-sectional — a single fasting-morning snapshot across people who differ in a hundred other ways — so it shows a correlation, not a dial you can turn.
Three designs, three ghrelin rises, from a two-night crash to one sleepless night to a population average. That is about as consistent as this field gets, which is why ghrelin is the hormone the popular story gets least wrong.
Leptin: the hormone that won't pick a direction#
Leptin is where the tidy narrative comes apart, and the disagreement is real rather than statistical noise. Spiegel's severe two-night restriction dropped it 18%. Taheri's cross-sectional data has it 15.5% lower in short sleepers. But Schmid's single sleepless night didn't move it at all. And a six-week trial of mild restriction — trimming just 78 minutes a night — found leptin going the other way, rising 2.03 ng/mL (Zuraikat et al., 2026; the full trial is the subject of how sleep loss drives weight gain).
These are not four noisy measurements of one number. They are different exposures asking a hormone that answers to far more than sleep. Leptin is secreted by fat tissue and tracks energy balance over days, so a two-night lab crash, a single all-nighter, and six weeks of mild restriction are genuinely different questions — and leptin runs on its own daily rhythm, so the timing of a blood draw can flip the sign of the result. A signal that goes down, stays flat, or goes up depending on how you probe it is not the thing reliably driving people to eat — and in people carrying more fat, the fullness message can stop landing however high leptin runs, which is leptin resistance.
The pooled evidence confirms the mess. A meta-analysis of randomized sleep-restriction trials found no strong evidence that short sleep significantly changes mean leptin or ghrelin — while the same analysis found subjective hunger up 13.4 mm on a 100 mm scale and food intake up 252.8 kcal a day5. Read that pairing twice: the hunger and the eating are real and statistically solid; the two hormones blamed for them wash out in the pooling.
The extra hunger and the extra eating are real and measurable. The two hormones everyone blames wash out when you pool the trials — so the popular story has the right symptom and the wrong cause.
Cortisol: the evening climb that actually replicates#
If any hormone earns a firm place in the story, it is cortisol — the stress hormone your body releases on a strong daily rhythm, high in the morning and low late at night. Lose sleep and that nighttime low arrives late and shallow. Studied across a 32-hour block in healthy young men, cortisol the evening after sleep loss ran 37% higher following partial deprivation (P = 0.03) and 45% higher after a fully sleepless night (P = 0.003), with the quiet trough of cortisol secretion delayed by at least an hour4. An evening of elevated cortisol is a plausible bridge between short sleep and appetite, because cortisol pushes up blood glucose and has long been linked to central, abdominal fat storage — the specifically-visceral part of that story is in sleep and belly fat.
One axis over sits insulin, and it behaves more consistently than the appetite hormones do: sleep restriction lowers insulin sensitivity, with a pooled standardized effect of −0.70 in the same meta-analysis5. That is a metabolic problem rather than a hunger signal, and it gets its own accounting in sleep and insulin sensitivity. For appetite specifically, cortisol is the sturdier of the two stress-side hormones.
The cascade, audited#
Put the four side by side and the picture is not "everything moves the wrong way." It is one dependable riser, one dependable evening climb, one hormone that can't commit to a direction, and a pooled verdict that quietly undercuts the tidy version.
| Hormone | The popular claim | What the controlled evidence shows | How solid |
|---|---|---|---|
| Ghrelin (hunger) | Rises with sleep loss | +28% over two nights1; +22% after one3 | Fairly consistent — but pools to non-significant5 |
| Leptin (fullness) | Falls with sleep loss | −18%1; unchanged3; up over six weeks6 | Contested — the direction itself is unstable |
| Cortisol (stress) | Rises with sleep loss | +37% to +45% the next evening4 | The most reproducible of the four |
| Insulin sensitivity | Drops with sleep loss | Pooled SMD −0.705 | Consistent, but metabolic, not appetite |
No row of that table says sleep does nothing. What it says is subtler and more useful: the hormone with the neatest story — leptin — is the least dependable of the set, and the appetite people actually feel runs ahead of what any of these hormones can account for.
Why the hormones get the credit and the behaviour does the work#
Here is the through-line the numbers keep drawing. Sleep-restricted people reliably feel hungrier and reliably eat more — Zhu's pooled 253 extra calories a day, and considerably larger figures in the sharpest single trials. But the leptin-and-ghrelin swing is too small, too inconsistent, and in leptin's case too directionally confused to be carrying that load on its own. Something else is doing most of the pushing.
That something lives mainly in the brain's reward and control circuits, not the bloodstream. Fed a full breakfast first — so ordinary hunger was off the table — sleep-deprived men still bought 9% more calories in a mock supermarket, and their elevated ghrelin predicted none of what they chose; the scanner and blood-assay evidence is walked through in why you crave junk food when you're tired. The reward system runs hot while the braking system runs cold, and the hormones are more passenger than driver.
This also sharpens what sleep is not doing. It is not slowing your furnace — resting metabolism barely budges, and if anything rises slightly, which is the argument in does poor sleep slow your metabolism. What the hormonal and behavioural mess does cost you, in a calorie deficit, is the quality of the loss: it steers weight off lean tissue instead of fat, covered in how sleep decides fat loss versus muscle loss, while the cortisol-driven drag on repair is the subject of why sleep is when muscle rebuilds. Whether all of it adds up to real weight over months is the cluster's headline question, in sleep and weight loss.
The practical translation is deflating on purpose. There is no hormone panel you can order, and no supplement that "resets leptin," that substitutes for the two moves with actual trial support: getting nearer to seven hours most nights if you're running short, and keeping protein high enough to blunt the appetite a bad night creates. Habitual short sleepers coached toward normal sleep ate less without adjusting a single hormone deliberately. The chemistry is a weather report; how much you sleep and what you eat is the climate.
FAQ#
Does sleep loss really lower leptin and raise ghrelin?#
Ghrelin, usually yes; leptin, not dependably. The famous experiment found ghrelin up 28% and leptin down 18% after two 4-hour nights1, but a single sleepless night raised ghrelin while leaving leptin unchanged3, and a six-week trial of mild restriction found leptin rising instead. Pooled across randomized trials, neither hormone shows a strong, significant mean change5.
If the hormones barely move, why am I so much hungrier when I'm tired?#
Because most of the effect is neural, not hormonal. The same pooled analysis that found little leptin or ghrelin change found subjective hunger clearly elevated and daily intake up about 253 kcal5. The driver is a reward system that runs hotter and a control system that runs weaker on short sleep, detailed in sleep and cravings — with the hunger hormones more passenger than pilot.
Can a supplement or a "leptin reset" fix sleep-related hunger hormones?#
No. There is no validated supplement that corrects sleep-related leptin or ghrelin, and given how inconsistently those hormones move with sleep in the first place, a product promising to "reset" one is selling a mechanism the evidence does not support. The two levers with real trial backing are unglamorous: sleep longer if you're short — habitual short sleepers who added about 1.2 hours ate measurably less — and eat enough protein to manage appetite directly.
Sources#
- Spiegel K, Tasali E, Penev P, Van Cauter E. Brief communication: sleep curtailment in healthy young men is associated with decreased leptin levels, elevated ghrelin levels, and increased hunger and appetite. Ann Intern Med. 2004;141(11):846-850.
- Taheri S, Lin L, Austin D, Young T, Mignot E. Short sleep duration is associated with reduced leptin, elevated ghrelin, and increased body mass index. PLoS Med. 2004;1(3):e62.
- Schmid SM, Hallschmid M, Jauch-Chara K, Born J, Schultes B. A single night of sleep deprivation increases ghrelin levels and feelings of hunger in normal-weight healthy men. J Sleep Res. 2008;17(3):331-334.
- Leproult R, Copinschi G, Buxton O, Van Cauter E. Sleep loss results in an elevation of cortisol levels the next evening. Sleep. 1997;20(10):865-870.
- Zhu B, Shi C, Park CG, Zhao X, Reutrakul S. Effects of sleep restriction on metabolism-related parameters in healthy adults: a comprehensive review and meta-analysis of randomized controlled trials. Sleep Med Rev. 2019;45:18-30.
- Zuraikat FM, Scaccia SE, Cochran JA, et al. Prolonged short sleep and its effect on body weight and composition: a pooled analysis of randomized trials. Ann Intern Med. 2026.
