Background: Insulin resistance is a causal factor in prediabetes (PD) and type 2 diabetes (T2D) and increases the risk of developing Alzheimer disease (AD). Reductions in cerebral glucose metabolic rate (CMRglu) as measured by fludeoxyglucose F 18-positron emission tomography (FDG-PET) in parietotemporal, frontal, and cingulate cortices are associated with increased AD risk and can be observed years before dementia onset. Objectives: To examine whether greater homeostasis model assessment insulin resistance (HOMA-IR) is associated with reduced resting CMRglu in areas vulnerable in AD in cognitively normal adults with newly diagnosed PD or T2D(PD/T2D), and to determine whether adults with PD/T2D have abnormal patterns of CMRglu during a memory encoding task. Design: Randomized crossover design of resting and activation FDG-PET. Setting: University imaging center and Veterans Affairs clinical research unit. Participants: Twenty-three older adults (mean [SEM] age, 74.4 [1.4] years) with no prior diagnosis of diabetes but who met American Diabetes Association glycemic criteria for PD (n=11) or diabetes (n=12) based on fasting or 2-hour oral glucose tolerance test (OGTT) glucose values and 6 adults (mean [SEM] age, 74.3 [2.8] years) with normal fasting glucose values and glucose tolerance. No participant met Petersen criteria for mild cognitive impairment. Interventions: Fasting participants underwent resting and cognitive activation FDG-PET imaging on separate days. Following a 30-minute transmission scan, subjects received an intravenous injection of 5 mCi of FDG, and the emission scan commenced 40 minutes after injection. In the activation condition, a 35-minute memory encoding task was initiated at the time of tracer injection. Subjects were instructed to remember a repeating list of 20 words randomly presented in series through earphones. Delayed free recall was assessed once the emission scan was complete. Main Outcome Measures: The HOMA-IR value was calculated using fasting glucose and insulin values obtained during OGTT screening and then correlated with CMRglu values obtained during the resting scan. Resting CMRglu values were also subtracted from CMRglu values obtained during the memory encoding activation scan to examine task-related patterns of CMRglu. Results: Greater insulin resistance was associated with an AD-like pattern of reduced CMRglu in frontal, parietotemporal, and cingulate regions in adults with PD/T2D. The relationship between CMRglu and HOMA-IR was independen to fage, 2-hour OGTT glucose concentration, or apolipoprotein E epsilon 4 allele carriage. During the memory encoding task, healthy adults showed activation in right anterior and inferior prefrontal cortices, right inferior temporal cortex, and medial and posterior cingulate regions. Adults with PD/T2D showed a qualitatively different pattern during the memory encoding task, characterized by more diffuse and extensive activation, and recalled fewer items on the delayed memory test. Conclusions: Insulin resistance may be a marker of AD risk that is associated with reduced CMRglu and subtle cognitive impairments at the earliest stage of disease, even before the onset of mild cognitive impairment.
