Low free testosterone is an independent risk factor for Alzheimer's disease
Introduction
Several recent studies have reported that low bio-available or free testosterone is associated with cognitive decline (Moffat et al., 2002, Yaffe et al., 2002) and Alzheimer's disease (AD) in men (Hogervorst et al., 2003, Moffat et al., 2004, Paoletti et al., 2004) and women (Paoletti et al., 2004). The fact that most studies reported low free (FT) but not total testosterone (TT) as a risk factor suggests that increased levels of sex hormone binding globulin (SHBG, which binds testosterone thus reducing bio-available levels) may play an important role in AD. SHBG is raised with age, in wasting syndrome, thyroid and liver dysfunction (Kaufman and Vermeulen, 1998) but it is unclear why it would be higher in cases with AD. Previous studies also did not determine whether the lower levels of testosterone had a primary (testicular) or secondary (hypothalamic-pituitary) origin. The gonadotropins—follicle-stimulating hormone (FSH) and luteinizing hormone (LH) produced by the pituitary gland—normally regulate sex steroid hormone production. While higher levels of gonadotropins in men with AD were reported (Bowen et al., 2000), this was no longer the case in another cohort when the analyses were adjusted for age (Short et al., 2001). No overall difference was found in our earlier cross-sectional study, but older (>73 years of age) AD cases did seem to have higher gonadotropin levels than older controls (Hogervorst et al., 2003). In the present study we increased the number of AD cases included (from n=45 to n=112) and did repeated measures to investigate why SHBG was raised in cases and how this related to gonadotropin levels in men with and without AD. As markers for thyroid function we included TSH and T4, as an indication of liver dysfunction we included albumin and alcohol abuse and as a marker of wasting syndrome we included body mass index (BMI) and creatinine. We controlled for diabetes mellitus as insulin has been known to affect free testosterone levels (Kaufman and Vermeulen, 1998).
Section snippets
Subjects
The Oxford Project To Investigate Memory and Ageing (OPTIMA) started in 1988 and has since included over 600 subjects. For the present study, serum samples of n=210 male participants over the age of 46 years were available. AD cases (n=112) and age-matched controls (n=98) had been followed up yearly with a full assessment, which included a physical examination (including weight and height measurements from which BMI was derived), blood tests, a detailed history corroborated by a close relative
Results
Mann–Whitney analyses showed that cases overall had significantly higher log FSH and slightly higher creatinine levels than controls. Trend differences (P<0.10) were seen for higher log LH and lower FAI, TT and TSH in the AD cases (Table 1).
There was no overall difference in age between AD cases and controls (average age 73.8 SD=9.2, P=0.73). GLM analyses including age showed that AD cases had significantly higher levels of log FSH [F(1,206)=4.65, P=0.03] and lower levels of FAI [F(1,921
Discussion
In cross sectional analyses, we found that elderly AD cases had higher levels of FSH than controls, but this was not apparent in the younger cases. For LH there were trends in a similar direction. Although FSH does not affect testosterone production directly, it has strong associations with LH levels. FSH is a more robust measure, has a longer half life and shows less circadian fluctuations than LH and could therefore be expected to be less sensitive to different types of confounds (slight
Acknowledgements
We would like to thank T. James and M. Gales at the Clinical Biochemistry Department of the John Radcliffe Infirmary for the testosterone, LH, FSH and SHBG assays. Richard Bowen made very helpful suggestions after reading a first draft of the paper. We would also like to thank the members and participants of OPTIMA for making this study possible. This work was supported by grants from Research Into Ageing/Help the Aged, Voyager Pharmaceutical, The Alzheimer's Association (NIRG 00-2258), the
References (23)
- et al.
The nature of the effect of female gonadal hormone replacement therapy on cognitive function in post-menopausal women: a meta-analysis
Neuroscience
(2000) - et al.
Analytical evaluation of the testosterone assay on the Bayer Immuno 1 system
Clin. Biochem.
(1998) - et al.
Elevated gonadotropin levels in patients with Alzheimer disease
Mayo. Clin. Proc.
(2001) - et al.
Apolipoprotein E genotyping by one-stage PCR
Lancet
(1991) - et al.
An association of elevated serum gonadotropin concentrations and Alzheimer disease?
J. Neuroendocrinol.
(2000) - et al.
Variations of sex hormone-binding globulin in thyroid dysfunction
Thyroid
(1999) - et al.
Folate vitamin B12 and serum homocysteine as candidate risk factors for confirmed Alzheimer's disease
Arch. Neurol.
(1998) Role of testosterone in maintaining lean body mass and bone density in HIV-infected patients
Int. J. Impot. Res.
(2003)- et al.
Testosterone and gonadotropin levels in men with dementia
Neuroendocrinol. Lett.
(2003) - et al.
Diagnosing dementia: interrater-reliability assessment and accuracy of the NINCDS/ADRDA criteria versus CERAD histopathological criteria for Alzheimer's disease
Dement. Geriatr. Cogn. Disord.
(2000)