Objectives: The purpose of this study was to determine whether the level of serum total testosterone (TT) was different in cases of Dementia of the Alzheimer's Type (DAT) than in controls.
Setting and design: We included 83 referred DAT cases and 103 cognitively screened volunteers (aged 75+/-9 years) from the Oxford Project To Investigate Memory and Ageing.
Methods: Information was obtained about potential confounds in the relation of DAT with testosterone, including age, gender, education, body mass index, smoking, (ab)use of alcohol, diabetes mellitus, endocrine therapy, and having undergone hysterectomy. TT was measured in non-fasting serum obtained between 10 and 12 a.m. using a competitive enzyme immunoassay.
Results: Men with DAT (n=39) had lower levels (p =0.005) of total serum testosterone (TT=14+/-5 nmol/L) than controls (n=41, TT=18+/-6 nmol/L). Lower TT was more likely in men with DAT, independent of potential confounds (Odds Ratio=0.78, 95% C.I.=0.68 to 0.91). In women there was no difference in TT levels between cases (n=44) and controls (n=62).
Main findings: Our results suggested that low TT may be a co-morbid feature of DAT in men. However, low TT levels could also exacerbate the disease.
Conclusions: Prospective longitudinal studies should investigate whether low TT levels precede or follow the onset of DAT (209 words).