Associations between testosterone and pre-androgens and sexual function; findings from the Australian Women's Midlife Years Study.

Abstract

Our aim was to determine the associations between these sex steroids and sexual function and difficulties in a national, community-based sample of women aged 40 to 69 years according to menopausal status. This was a prespecified sub-study of the cross-sectional Australian Women's Midlife Years Study. Participants not pregnant, breastfeeding, on medications affecting sex hormone concentrations, or living over 100km from blood collection centres were invited to provide a blood sample. Participants with thyroid dysfunction, hyperprolactinemia, pregnancy, bilateral oophorectomy, indiscernible menopause stage, missing samples, moderate-severe depressive symptoms, or psychotropic medication use were excluded from the analyses. Sex hormones measured by liquid chromatography and tandem mass spectrometry. The Profile of Female Sexual Function questionnaire completed between 27October 2023 and 19March 2024. The analysis included 136 premenopausal and 595 perimenopausal and postmenopausal (peri-postmenopausal) participants, with median ages (range) of 45 (40-57) years and 61 (40-69) years, respectively. After adjustment for age, BMI, residential location, ancestry, education, relationship status, smoking, alcohol, LGBTQIA+, moderate to severe vaginal dryness, and ever sexual abuse, no association was found between any hormone and sexual desire, or desire difficulty. Testosterone (cubic p=0.013) and androstenedione (cubic p=0.013) were nonlinearly associated with orgasm in premenopausal and peri-postmenopausal, respectively, and androstenedione with arousal in both groups (cubic p=0.028 and p=0.011, respectively). Peri-post menopausal women with orgasm difficulty had lower median [IQR] serum testosterone (0.39 [0.31, 0.55] vs 0.48 [0.33, 0.68] nmol/L, p=0.025), DHEA (5.11 [3.29, 7.56] vs 6.17 [4.33, 8.25] nmol/L, p=0.012) and androstenedione (1.06 [0.86, 1.46] vs 1.29 [0.91, 1.85], p=0.038). Testosterone and androstenedione were lower in peri-postmenopausal women with responsiveness difficulty (0.39 [0.29, 0.53] vs 0.47 (0.33, 0.68) nmol/L, p=0.012 and 1.08 (0.82, 1.46) vs 1.28 (0.91, 1.84) nmol/L, p=0.006, respectively). Testosterone and the pre-androgens were not associated with sexual desire in midlife women. The associations between sex steroids and orgasm and arousal were complex, and varied with menopausal status. Measurement of these sex steroids should not be part of the routine clinical assessment of women presenting with sexual concerns.