SHBG, Free Testosterone, and Type 2 Diabetes Risk in Middle-aged African Men: A Longitudinal Study

• Published in: Journal of the Endocrine Society Vol. 8; no. 8; p. bvae129
• Main Authors: Seipone, Ikanyeng D, Mendham, Amy E, Storbeck, Karl-Heinz, Oestlund, Imken, Kufe, Clement N, Chikowore, Tinashe, Masemola, Maphoko, Crowther, Nigel J, Kengne, Andre Pascal, Norris, Shane, Olsson, Tommy, Brown, Todd, Micklesfield, Lisa K, Goedecke, Julia H
• Format: Journal Article
• Language: English
• Published: United States Oxford University Press 01.07.2024
• Subjects: Africa; Albumin; Clinical; Dextrose; dysglycaemia; free testosterone; Glucose; Glucose metabolism; Glucose tolerance tests; HIV (Viruses); Insulin; Insulin resistance; Physiological aspects; Risk factors; sex hormone-binding globulin; Testosterone; Type 2 diabetes; South Africa; Sub-Saharan Africa; free testosterone; type 2 diabetes; sex hormone-binding globulin; dysglycaemia; Africa
• ISSN: 2472-1972; 2472-1972
• DOI: 10.1210/jendso/bvae129

To investigate longitudinal changes in SHBG and free testosterone (free T) levels among Black middle-aged African men, with and without coexistent HIV, and explore associations with incident dysglycaemia and measures of glucose metabolism. This longitudinal study enrolled 407 Black South African middle-aged men, comprising primarily 322 men living without HIV (MLWOH) and 85 men living with HIV (MLWH), with normal fasting glucose at enrollment. Follow-up assessments were conducted after 3.1 ± 1.5 years. At baseline and follow-up, SHBG, albumin, and total testosterone were measured and free T was calculated. An oral glucose tolerance test at follow-up determined dysglycaemia (impaired fasting glucose, impaired glucose tolerance, type 2 diabetes) and glucose metabolism parameters including insulin sensitivity (Matsuda index), insulin resistance (homeostasis model assessment of insulin resistance), and beta(β)-cell function (disposition index). The primary analysis focussed on MLWOH, with a subanalysis on MLWH to explore whether associations in MLWOH differed from MLWH. The prevalence of dysglycaemia at follow-up was 17% (n = 55) in MLWOH. Higher baseline SHBG was associated with a lower risk of incident dysglycaemia (odds ratio 0.966; 95% confidence interval 0.945-0.987) and positively associated with insulin sensitivity (β = 0.124, < .001) and β-cell function (β = 0.194, = .001) at follow-up. Free T did not predict dysglycaemia. In MLWH, dysglycaemia prevalence at follow-up was 12% (n = 10). Neither baseline SHBG nor free T were associated with incident dysglycaemia and glucose metabolism parameters in MLWH. SHBG levels predict the development of dysglycaemia in middle-aged African men but do not exhibit the same predictive value in MLWH.