MK-677 and Sleep: Does Ibutamoren Improve Sleep Quality?

Of all the growth-hormone peptides marketed for better sleep, MK-677 (ibutamoren) has something the others don't: an actual polysomnography trial. Where sermorelin and ipamorelin ride on the general growth-hormone-and-deep-sleep rationale, MK-677 was measured in a sleep lab, and the wires recorded more deep sleep and more REM. That makes the MK-677 sleep claim the best-supported one in this whole category — which is exactly why it needs the most careful reading. "Best-supported" here means a single small crossover study from the late 1990s, run during the drug's own development, never replicated, and attached to a compound that never won approval and now trades grey-market with a meaningful metabolic downside. The evidence is more real than the marketing around sermorelin, and more fragile than the marketing around MK-677. Both things are true.

What MK-677 is (and why sleep even comes up)

MK-677 is not a peptide. It's a small, orally active, non-peptide molecule that mimics the stomach hormone ghrelin at the growth hormone secretagogue receptor (GHS-R)⁴ — the receptor identified in the pituitary and hypothalamus in 1996⁴ whose natural ligand, ghrelin, was discovered in 1999⁵. Because it survives digestion, one nightly oral dose keeps growth hormone and IGF-1 elevated for many hours. In young men, seven days of MK-677 raised GH pulse frequency and IGF-1 while total 24-hour GH secretion was essentially unchanged²; in healthy elderly subjects, daily oral MK-677 restored the GH/IGF-1 axis toward young-adult levels³. That is settled — MK-677 reliably raises the hormone. The sleep question is whether that translates into better nights, and here the ghrelin receptor it hits is the interesting part, because ghrelin is itself woven into sleep regulation, not just growth hormone. (For how MK-677 stacks up against the injectable GHRH peptide this site is built around, see sermorelin vs MK-677, and for the two arms of the GH axis it sits on, GHRH vs GHRP.)

The one trial that actually measured sleep

The anchor is a 1997 study in Neuroendocrinology by Copinschi and colleagues — a double-blind, placebo-controlled, crossover trial that gave MK-677 at bedtime and recorded polysomnography in both young (18–30) and older (65–71) adults¹. The results are genuinely striking for a sleep aid: in young subjects, high-dose MK-677 produced roughly a 50% increase in stage IV (the deepest slow-wave) sleep and a more than 20% increase in REM sleep versus placebo, and the frequency of "deviations from normal sleep" fell from 42% on placebo to 8% on the drug¹. In the older adults, REM sleep rose by nearly 50% with a shortened REM latency, and their sleep disturbances dropped too¹. The authors concluded MK-677 might simultaneously improve sleep quality and correct the age-related decline in growth-hormone output¹.

Taken at face value, that is a better direct sleep result than anything published for sermorelin or ipamorelin, whose sleep case rests on GHRH and GH-axis studies rather than trials of the molecule itself (we hold that line in ipamorelin and sleep and sermorelin and deep sleep). So credit where due: MK-677 was actually wired up in a sleep lab, and it moved the deep-sleep and REM numbers.

Why one striking trial isn't proof

Now the fine print, because it's substantial. The trial was tiny — eight young subjects and six older ones — which is a normal size for an intensive sleep-lab study but far too small to treat as a settled result. It is a single study, run in the late 1990s during MK-677's clinical development, and it has not been replicated in the quarter-century since. Its endpoints are sleep architecture (how much stage IV and REM the EEG records) plus a disturbance count — meaningful, but not the same as a large trial showing people consistently feel more rested, function better the next day, or sleep better over months of real-world use. In evidence terms this is a promising signal from one small, old, unreplicated study, not an established effect. It deserves to be reported honestly as exactly that — neither dismissed nor inflated into a guarantee.

The mechanism is real but messier than it looks

There is a genuine biological reason a ghrelin-receptor drug might touch sleep. Ghrelin activates the somatotropic axis, and the largest natural growth-hormone pulse of the day happens during slow-wave sleep, so the two systems are coupled. Direct evidence backs part of this: infusing ghrelin promotes slow-wave sleep in humans⁶, which lines up neatly with the stage IV increase Copinschi saw. But the mechanism does not cleanly explain the whole result. When researchers infused ghrelin itself in young men, non-REM sleep rose but REM sleep actually fell⁷ — the opposite of the REM increase reported under MK-677¹. In other words, the deep-sleep (slow-wave) part of the MK-677 finding fits the ghrelin story, but the REM increase does not obviously follow from ghrelin-receptor activation, and may reflect the broader restoration of the GH/IGF-1 rhythm rather than a direct receptor effect. The honest read: the mechanism makes a deep-sleep benefit plausible, but it is not a tidy, fully-explained pathway, and the human sleep data on the ghrelin arm of the axis are themselves mixed.

The metabolic bill comes due

Here is where a sleep-and-healthy-aging reader has to weigh the whole package, not just the EEG. The best long-term human data on MK-677 is a two-year, randomized, double-blind, placebo-controlled trial in healthy older adults⁸. It confirmed the drug raises GH and IGF-1 to young-adult levels and increased fat-free mass — but the increased fat-free mass did not translate into better strength or function, body weight rose by about 2.7 kg, and, critically, fasting blood glucose increased and insulin sensitivity decreased⁸. Appetite increase and mild lower-limb edema were the most common side effects⁸. A safety review of the growth-hormone-secretagogue class flags the same pattern — the ghrelin-mimetic mechanism predictably drives hunger and worsened glucose handling¹⁰, and fluid retention (mild lower-limb edema) is documented in longer-term use⁸. And because MK-677 keeps IGF-1 elevated around the clock rather than in a short overnight pulse, those metabolic effects are harder to escape between doses.

For someone chasing sleep, recovery, and metabolically healthy aging, that trade is pointed the wrong way. Rising blood sugar, an appetite spike, and water retention are precisely the things that undercut the goals people come to this category for — and elevated overnight glucose can itself fragment sleep. So even granting the deep-sleep signal, MK-677 asks you to accept a documented metabolic cost to get it. Anyone with prediabetes, diabetes, or metabolic-syndrome risk should treat the glucose effect as the headline, not a footnote. (We cover the GH-axis glucose question more fully in sermorelin, blood sugar & diabetes.)

The supply chain undercuts the whole thing

One more caution that changes the calculus. MK-677 has never been an FDA-approved drug. Despite an extensive clinical program — for the somatopause, frailty, and hip-fracture recovery, where a randomized phase IIb trial did not lead to an approved product⁹ — it never earned an indication, and development was halted. Today it circulates as a "research chemical" sold for laboratory use and not for human consumption, and it appears on the World Anti-Doping Agency's prohibited list. Practically, that means a grey-market vial or capsule with no guarantee of identity, dose accuracy, or purity — a hazard entirely separate from the molecule's own effects. The sleep trial used pharmaceutical-grade MK-677 under medical supervision; a research-chemical purchase is not the same product under the same conditions. This is the sharp contrast with compounded sermorelin, which at least keeps a prescriber and a licensed pharmacy in the loop — we lay out that landscape in the pillar guide to sermorelin's evidence.

The bottom line on MK-677 and sleep

MK-677 is the rare growth-hormone secretagogue with a real sleep-lab trial behind it: a 1997 crossover study found bedtime dosing increased deep (stage IV) and REM sleep and cut sleep disturbances in both young and older adults¹. That is a stronger direct sleep signal than sermorelin or ipamorelin can claim. But it is one small, old, never-replicated study measuring sleep architecture, not long-term rested-ness; the ghrelin mechanism explains the deep-sleep part better than the REM part; and MK-677's own long-term data show it raises fasting glucose, spikes appetite, causes edema, and never won approval — so it now trades as an unregulated research chemical. If sleep quality and healthy aging are the goal, the sleep signal is real but comes bundled with a metabolic cost and a supply chain you can't verify. For the peptides more commonly used with sleep in mind, compare notes in our guide to peptides for sleep, and if you're weighing a prescriber-supervised route, we rank oversight and pricing in the best sermorelin providers.