Premium
Age and cognitive diagnosis influence cerebrospinal fluid ketone levels after a triglyceride infusion in older adults
Author(s) -
Hanson Angela J.,
Banks William A.,
Bettcher Lisa F.,
Pepin Robert,
Raftery Daniel,
Craft Suzanne
Publication year - 2020
Publication title -
alzheimer's and dementia
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 6.713
H-Index - 118
eISSN - 1552-5279
pISSN - 1552-5260
DOI - 10.1002/alz.037716
Subject(s) - ketone bodies , medicine , triglyceride , endocrinology , saline , cerebrospinal fluid , cognitive decline , chemistry , cholesterol , metabolism , dementia , disease
Background Ketone bodies and diets which increase plasma ketones are being investigated for treatments for Alzheimer’s disease (AD), but little is known about what influences ketone body levels in the brain. Method Older adults (n=21, age 67.7 ± 8.6) underwent a 5‐hour TG and saline infusion on different days in random crossover design; CSF was collected at the end of the infusion. Targeted aqueous metabolites were measured using Shimadzu Nexera 20/AB‐Sciex 6500 Triple Quadrupole Liquid Chromatography‐Mass Spectroscopy which targets 216 metabolites representing over 35 different metabolic pathways. Data were analyzed using MetaboAnalyst 4.0 and SAS. Result Of 216 possible metabolites, 99 were detectable in the CSF, with an average of 88 metabolites per sample which did not differ by treatment (saline 87.6 ± 1.4, TG 87.7 ± 1.7). Three metabolites significantly differed by treatment: 3‐hydroxybutyrate (HBA), glutamic acid and methylguanosine (Table). The top hit HBA further differed by cognitive diagnosis. Fasting HBA was similar for both groups but TG‐induced increases in HBA were over 3 times higher for those with cognitive impairment (change score normal control 9.8 uM ± 8.3, cog impair 32.4 uM ±7.4, p value log transformed levels 0.0198). None of the metabolites in Table differed by APOE status or by sex. Age positively correlated with fasting HBA whereas systolic blood pressure and fasting glucose positively correlated with TG HBA and HBA change score. Fasting lipid levels did not predict fasting or TG‐induced HBA. Conclusion We detected metabolites in human CSF, and most metabolites did not change after a TG infusion. The top metabolite was the ketone body HBA, and levels were influenced by age, cognitive diagnosis, and markers of the metabolic syndrome. These results suggest that interventions that increase plasma ketones may lead to higher brain ketones in groups at risk for AD.