Sleep disturbances affect up to half of older adults with type 2 diabetes mellitus (T2DM) and represent a clinically meaningful but under addressed dimension of diabetes care. Metformin, the universally recommended first-line agent for T2DM, exerts multiple pleiotropic effects, including modulation of AMPK signaling, circadian clock gene expression, neuroinflammation, and neuropathic pain, that are each mechanistically relevant to sleep regulation. Although metformin may influence pathways relevant to sleep, the evidence linking metformin use to sleep outcomes in older adults with T2DM has not been systematically reviewed. This narrative review was conducted in accordance with the SANRA framework, drawing on a systematic search of PubMed, Scopus, and Web of Science covering literature published between 2008 and 2025. Studies were included if they reported sleep outcomes in T2DM patients aged 60 years or older involving metformin, or if they provided mechanistic evidence relevant to the metformin-sleep interface. Both subjective and objective sleep measures were considered alongside mechanistic, pharmacological, and safety evidence. Metformin's pleiotropic pharmacology converges on four biologically plausible pathways through which sleep may be improved in elderly T2DM patients: glycemic stabilization reducing nocturia, AMPK-dependent attenuation of neuroinflammation, modulation of circadian clock gene expression, and neuroprotective effects preserving sleep-regulating neural circuits. Available clinical evidence suggests modest improvements in sleep-related outcomes, particularly among patients with Pittsburgh Sleep Quality Index scores above 5, although observational study designs and the absence of dedicated randomized trials limit causal inference. Rare adverse sleep effects, including vivid dreams and nightmares, appear uncommon and generally resolve with dose adjustment. Further randomized controlled trials are warranted to determine the effects of metformin on sleep outcomes in elderly patients with T2DM.
