Ketones and Sleep: A Narrative Review of Mechanisms, Preclinical Models, and Human Evidence

Background/Objectives: Sleep is a metabolically dynamic state, and ketone bodies may interact with sleep physiology as both oxidative substrates and signaling metabolites. β-Hydroxybutyrate (BHB), acetoacetate (AcAc), and acetone rise during fasting, carbohydrate restriction, ketogenic diets used for therapeutic or non-therapeutic purposes, prolonged exercise, selected pathological states of insulin deficiency, and after consumption of exogenous ketones. This narrative review synthesizes mechanistic, preclinical, and human evidence linking endogenously generated ketosis and ketosis generated from exogenous ketone consumption to sleep architecture, sleep quality, sleep-disordered breathing, and sleep-relevant metabolic pathways. Methods: We reviewed studies involving ketogenic diets, fasting or caloric restriction, and consumption of exogenous ketones, prioritizing investigations with polysomnography, electroencephalography (EEG), sleep architecture, respiratory, or validated sleep-questionnaire outcomes, while also considering mechanistic studies of neuronal excitability, inflammation, epigenetic regulation, circadian timing, and brain energetics. Results: The most direct preclinical evidence indicates that AcAc, but not BHB, increases slow-wave activity (delta power 0.5–4 Hz) during non-rapid eye movement (NREM) sleep and suppresses glutamate release in mice, supporting a potential role for ketone metabolism in sleep homeostasis. Dietary, fasting, and aging models further suggest that lipid/ketone metabolism can alter NREM/rapid eye movement (REM) balance, circadian metabolic rhythms, and sleep–wake responses to energetic stress, with aged-rat data showing sex-specific REM and cognitive effects of ketogenic interventions. Human studies are suggestive but heterogeneous: acute carbohydrate restriction can increase slow-wave sleep and reduce REM sleep in healthy adults; ketogenic therapies may improve subjective sleep in selected neurological and metabolic populations; and ketone ester ingestion can improve sleep efficiency after strenuous evening exercise, but not during hypoxia-induced sleep disruption. Conclusions: Ketones are sleep-relevant metabolic signals, but their effects are conditional rather than universally sleep-promoting. They may also indirectly affect sleep by enhancing wakefulness quality. The ketone concentration range associated with optimal sleep outcomes remains undefined, and potential dose-, timing-, and context-dependent effects on sleep require further study. Future studies should distinguish AcAc from BHB biology, measure nocturnal ketone dynamics, incorporate PSG/EEG spectral endpoints, and separate ketosis-specific effects from weight loss, energy deficit, exercise load, respiratory physiology, circadian timing, and disease-related confounders.