Insulin and cognitive function in humans: experimental data and therapeutic considerations

Abstract

Data from experimental studies in animals and from epidemiological studies in humans suggest a link between insulin and cognitive performance. Do these results translate into clinical and therapeutic benefit for people with cognitive impairment? Insulin injected peripherally can readily cross the blood–brain barrier. Intravenous insulin can improve aspects of cognitive function in healthy adults and in individuals with Alzheimer's dementia. Moreover, intravenous insulin increases concentrations of a long form of β-amyloid protein, Aβ42. One potential confounding factor with these data, however, is the need for co-administration of glucose with the insulin to maintain euglycaemia as glucose itself can facilitate memory function. Administration of insulin via the intranasal route is scientifically (and therapeutically) more attractive because the insulin goes directly to the cerebrospinal fluid, with minimal systemic absorption; this obviates the need for a glucose infusion. Intranasal insulin may improve some aspects of memory in healthy individuals, but has yet to be studied in people with cognitive impairment. TZDs (thiazolidinediones) reduce peripheral insulin concentrations by enhancing insulin sensitivity. In adults with Type II (non-insulin-dependent) diabetes, TZD therapy improves memory function, but so does sulphonylurea therapy (which elevates peripheral insulin concentrations). Improved memory is linked to lower blood glucose concentrations, rather than altered insulin levels. However, major trials are currently under way examining the impact of TZDs in people with dementia.