OR12-1 Differential associations of plasma testosterone, dihydrotestosterone and estradiol measured using liquid chromatography-tandem mass spectrometry with all-cause mortality in a population-based cohort of older men

Program: Abstracts - Orals, Featured Poster Presentations, and Posters
Session: OR12-Reproductive Hormones: Vital Effects Evidenced In Human Cohorts & Experimental Models
Saturday, June 15, 2013: 11:30 AM-1:00 PM
Presentation Start Time: 11:30 AM
Room 102 (Moscone Center)
Bu Beng Yeap*1, Helman Alfonso2, Paul Chubb3, David J Handelsman4, Graeme J Hankey2, Jonathan Golledge5 and Leon Flicker6
1University of Western Australia, Fremantle, Australia, 2University of Western Australia, 3Fremantle Hospital, Fremantle, Australia, 4University of Sydney, Sydney NSW, Australia, 5James Cook University, Townsville, Australia, 6University of Western Australia, Perth, Australia

In men testosterone (T) levels decline with increasing age, and previous studies have found that lower T levels predict increased mortality in older men. Whether reduced circulating T contributes to poorer health outcomes or reflects the presence of pre-existing disease remains debated. Furthermore the associations of its metabolites dihydrotestosterone (DHT) and estradiol (E2) with mortality are poorly defined.


We assessed the associations of circulating T, DHT and E2 with all-cause mortality in older men.


Community-dwelling men aged 70-89 years resident in Perth, Western Australia.

Main outcome measures

Plasma T, DHT and E2 were assayed using liquid chromatography-tandem mass spectrometry in early morning samples collected in 2001-04 from 3,690 men. Sex hormone-binding globulin (SHBG) and luteinising hormone (LH) were measured by immunoassay. Deaths to 31 December 2010 were ascertained using the Western Australian Data Linkage System. Cox proportional hazards regression was performed to assess associations of hormones with all-cause mortality. Adjustments were made for age, education, smoking, body mass index, waist:hip ratio, hypertension, dyslipidemia, diabetes, creatinine, cancer and cardiovascular disease.


There were 974 deaths (26.4%) at a median follow-up of 7.1 years. Men who died had lower baseline levels of T (mean±SD: 12.8±5.1 vs 13.2±4.8 nmol/L, p=0.013), DHT (1.4±0.7 vs 1.5±0.7 nmol/L, p=0.002) and E2 (71.6±29.3 vs 74.0±29.0 pmol/L, p=0.022) compared to surviving men. After adjusting fully for covariates, there were U-shaped associations of T (reference lowest quartile, Q1: hazard ratio [HR] for Q2=0.82, 95% confidence interval [CI]=0.69-0.98, p=0.033; HR for Q3=0.78, 95% CI= 0.65-0.94, p=0.010) and DHT (reference Q1: HR for Q3=0.76, 95% CI=0.63-0.91, p=0.003) with all-cause mortality. E2 was not associated with mortality in the regression analysis. When either SHBG or LH were included with T in multivariable models, the association of T with death from any cause remained significant.


Older men with mid-range levels of T and DHT exhibited the lowest rates of death from any cause, after adjustment for other risk factors. T predicted mortality independently of LH and SHBG. Optimal androgen exposure may be a contributing factor or robust biomarker for survival. Interventional studies are needed to clarify whether modulating T levels would improve male longevity.

Nothing to Disclose: BBY, HA, PC, DJH, GJH, JG, LF

*Please take note of The Endocrine Society's News Embargo Policy at http://www.endo-society.org/endo2013/media.cfm

Sources of Research Support: BBY is recipient of a Clinical Investigator Award from the Sylvia and Charles Viertel Charitable Foundation, New South Wales, Australia. The Health In Men Study was funded by Project Grants 279408, 379600, 403963, 513823 and 634492 from the National Health and Medical Research Council of Australia. JG is supported by fellowships and grants from the NHMRC, Queensland Government and The BUPA foundation. The funding sources had no involvement in the planning, analysis and writing of the manuscript.
Previous Abstract | Next Abstract >>