Relationship of metabolic and endocrine parameters to brain glucose metabolism in older adults: do cognitively-normal older adults have a particular metabolic phenotype?

Abstract

Our primary objective in this study was to quantify whole brain and regional cerebral metabolic rates of glucose (CMRg) in young and older adults in order to determine age-normalized reference CMRg values for healthy older adults with normal cognition for age. Our secondary objectives were to--(i) report a broader range of metabolic and endocrine parameters including body fat composition that could form the basis for the concept of a 'metabolic phenotype' in cognitively normal, older adults, and (ii) to assess whether medications commonly used to control blood lipids, blood pressure or thyroxine affect CMRg values in older adults. Cognition assessed by a battery of tests was normal for age and education in both groups. Compared to the young group (25 years old; n = 34), the older group (72 years old; n = 41) had ~14% lower CMRg (μmol/100 g/min) specifically in the frontal cortex, and 18% lower CMRg in the caudate. Lower grey matter volume and cortical thickness was widespread in the older group. These differences in CMRg, grey matter volume and cortical thickness were present in the absence of any known evidence for prodromal Alzheimer's disease (AD). Percent total body fat was positively correlated with CMRg in many brain regions but only in the older group. Before and after controlling for body fat, HOMA2-IR was significantly positively correlated to CMRg in several brain regions in the older group. These data show that compared to a healthy younger adult, the metabolic phenotype of a cognitively-normal 72 year old person includes similar plasma glucose, insulin, cholesterol, triglycerides and TSH, higher hemoglobin A1c and percent body fat, lower CMRg in the superior frontal cortex and caudate, but the same CMRg in the hippocampus and white matter. Age-normalization of cognitive test results is standard practice and we would suggest that regional CMRg in cognitively healthy older adults should also be age-normalized.