CORTISOL, DEHYDROEPIANDROSTERONE (DHEA), AND DHEA SULFATE IN THE CEREBROSPINAL-FLUID OF MAN - RELATION TO BLOOD-LEVELS AND THE EFFECTS OF AGE

The relation between blood and cerebrospinal fluid (CSF) concentration s of cortisol, dehydroepiandrosterone (DHEA), and its sulfate (DHEAS) was measured in 62 subjects aged 3-85 yr old, fitted with ventriculo-p eritoneal or lumbar-peritoneal shunts for a variety of diagnoses. Ther e were 36 males and 36 females. Forty-eight subjects were not taking e xogenous corticosteroids; the other 14 were receiving either systemic or local steroids. A single sample of blood and CSF was taken from eac h subject within 10 min for measurement of cortisol, DHEA, and DHEAS. The proportional levels of cortisol (5.8%) and DHEA (5.4%) in the CSF compared with those in the blood were similar in subjects not taking s teroids. However, CSF DHEAS levels were only 0.15% of those in the blo od. Because DHEAS blood levels were so much greater than DHEA, DHEAS i n the CSF was still higher than DHEA despite the reduced penetration o f the sulfated steroid. The blood/CSF ratios were similar in subjects taking steroids. There were significant correlations in steroid-free s ubjects between blood and CSF levels for DHEA (r = 0.65) and DHEAS (r = 0.88) but not for cortisol (r = 0.26). Steroid treatment significant ly lowered blood cortisol, DHEA and DHEAS, and CSF DHEA, but not CSF c ortisol or DHEAS compared with an age- and sex-matched sample of stero id-free subjects. In steroid-free adults (18 yr and over; n = 37), blo od cortisol showed no age-related change. However, CSF cortisol was ma rkedly raised in a proportion of steroid-free subjects over the age of 60 yr. Levels of corticoid-binding globulin in plasma did not alter w ith age. As expected, there were significant age-related decrements in both blood DHEA and DHEAS. CSF DHEA(r = 0.42) and CSF DHEAS (r = 0.39 ) were significantly negatively correlated with age. In steroid-free j uveniles (n = 11) there were no age-related changes in either blood or CSF cortisol, but significant increases with age in DHEA and DHEAS in both blood and CSF. Calculation of the cortisol/DHEA and cortisol/DHE AS molar ratios in the CSF showed both to be raised in the very young (3-8 yr) and the elderly (60 yr and over) by a factor of 4-5 compared with young adults aged 18-39. There were no sex differences in any of the parameters measured. These findings show that the relation between levels in the blood and CSF differ for each of these three neuroactiv e steroids. The brain is exposed to relatively high levels of DHEA and DHEAS during later childhood and early adulthood but to relatively or absolutely high levels of cortisol during infancy and older age. In v iew of the known antiglucocorticoid action of DHEA and DHEAS, and the direct action of these steroids on membrane-bound transmitter events ( such as gamma-aminobutyric acid, receptors), these changes may have im portant implications for age-related alterations in brain function.