Androgen Deficiency, Associations and Survival of Men With Stage 4 and 5 Chronic Kidney Disease: A Cohort Study

Objectives: Anaemia is a key cause of morbidity in chronic kidney disease (CKD). Androgen deficiency (AD) in males can contribute to anaemia of all causes, including in CKD. We sought to examine the prevalence of AD in men with CKD, the extent to which it contributed to anaemia and whether it was independently associated with long-term survival.

Methods: This cross-sectional observational study was conducted among males aged 18 years and over with CKD stages 4 and 5. The study analysed morning blood samples with regard to their full blood count, urea and electrolytes, albumin, lipids, testosterone (T) and sex hormone binding globulin, with calculation of free testosterone by mass action equation. Mortality data were obtained 15 years later for survival analysis.

Results: Among 322 patients with a mean age of 63 years, the overall prevalence of AD was 68.9%. There was a statistically significant negative correlation between erythropoiesis stimulating agent (ESA) dose and testosterone concentrations (Pearson correlation -0.193, p = 0.05). There was a positive correlation between haemoglobin (Hb) and free testosterone level among patients not on ESA therapy (Pearson correlation 0.331, p < 0.001). Kaplan-Meier plots showed p < 0.001 on log-rank analysis, indicating that AD was significantly associated with worse survival. However, in Cox regression analysis, free testosterone was not associated with survival (95% CI for free testosterone 0.997-1.000).

Conclusions: AD is highly prevalent among this population, and increases further with older age and more severe CKD warranting haemodialysis. Association of lower Hb and higher ESA dose with lower T concentration might be causative, which has important pharmaco-economic as well as clinical implications. Lower survival in men with low T, more likely reflects overall poor health rather than causation. A properly constituted randomised controlled study evaluating the effect of native T replacement is warranted in men with CKD and AD.