Sleep-related metabolites in healthy individuals: a systematic review with implications for cardiometabolic health

Background Poor sleep is strongly linked to cardiometabolic outcomes (1-3). Moreover, sleep was included as a key cardiovascular metric by the American Heart Association in 2022 (4), highlighting the need to understand underlying biological pathways. Metabolomics can uncover metabolic alterations associated with sleep, providing insight into pathways relevant to cardiovascular health. Purpose To systematically review and synthesize existing evidence linking different sleep phenotypes with alterations in key metabolites in healthy individuals without sleep disorders. Methods The systematic review protocol is registered in the International Prospective Register of Systematic Reviews (PROSPERO) network (5). A search was conducted in 6 databases (Cochrane, Embase, LILACS, PubMed, Scopus, WoS) using a combination of keywords (and synonyms) related to sleep and metabolomics (5). Data screening, extraction and quality assessment were performed independently by 2 investigators using the PRISMA guidelines. Sleep-related variables were grouped into predefined domains: short sleep duration, sleep-disordered breathing, sleep timing, insomnia, heart rate, excessive daytime sleepiness and multi-component sleep quality score. Metabolites reported in ≥2 studies with the same direction were considered associated with sleep. A summary of evidence score was constructed for each metabolite using a formula combining number of studies and consistency of direction: (increased−decreased)×log10*(total studies+1). Metabolites were classified as strong, medium, or weak evidence (|score| ≥65, ≥35, or <35, respectively). Results Of the 1059 screened records, 27 studies evaluating associations between sleep-related variables and metabolomics were included (Fig 1). We identified consistent metabolomic alterations across several sleep domains (Table 1). A total of 60 metabolites showed consistent associations: 5 with strong evidence, 34 with medium evidence (mainly increased; only 3 decreased), and 21 with weak evidence (13 increased and 8 decreased). Amino acids and their derivatives were predominantly increased with unhealthier sleep patterns. Isoleucine showed strong evidence of association, and its related metabolic γ-glutamyl dipeptides (γ-glutamylisoleucine and γ-glutamylleucine) demonstrated medium evidence with consistent increases, suggesting involvement of branched-chain amino-acid pathways. Phenylalanine also showed strong evidence. Lipid classes were similarly elevated with poor sleep habits, including phosphatidylcholines and lysophosphatidylcholines, with strong evidence for PC 32:1, LysoPC 14:0, and LysoPC 16:1, alongside medium-evidence increases in DAGs, TAGs, and free fatty acids. Conclusion Metabolites involved in amino acid metabolism, energy regulation, and lipid remodelling are associated with sleep, representing potential targets for future research into the biological pathways connecting sleep and cardiovascular risk.For image description, please refer to the figure legend and surrounding text. For image description, please refer to the figure legend and surrounding text.