Testosterone is lower in men with non-alcoholic fatty liver disease and alcohol-related cirrhosis and is associated with adverse clinical outcomes

SUMMARY

Background/aims

Low serum testosterone is common in cirrhotic men, but the impact of disease aetiology remains uncertain. This study compares serum total testosterone (TT) levels by disease aetiology and assesses its prognostic value.

Methods

Single-centre retrospective study of cirrhotic men who had TT levels measured between 2002 and 2020. A cut-off of 12 nmol/L was used to define low TT and 230 pmol/L for calculated free testosterone (cFT). Linear and logistic regression used to adjust for variables known to affect testosterone levels and assess for an association between levels and outcomes.

Results

Of 766 cirrhotic men, 33.3% had alcohol-related liver disease (ALD) and 11.9% had non-alcoholic fatty liver disease (NAFLD). The median age was 56 years (interquartile range (IQR) 50–61), and the model for end-stage liver disease (MELD) score 14 (IQR 9-20). TT levels were low in 53.3% of patients, (median 11.0 nmol/L; IQR 3.7–19.8) and cFT low in 79.6% (median 122 pmol/L; IQR 48.6-212). Median TT was lower in men with ALD (7.6 nmol/L; IQR 2.1–16.2) and NAFLD (9.8 nmol/L; IQR 2.75-15.6) compared to other aetiologies (11.0 nmol/L; IQR 3.73-19.8) (p < 0.001 for all), which remained true after adjustment for age and MELD score. TT was inversely associated with 12-month mortality or transplant (381 events, p = 0.02) and liver decompensation (345 events, p = 0.004).

Conclusions

Low serum testosterone is common in cirrhotic men and is associated with adverse clinical outcomes. TT levels are significantly lower in ALD and NAFLD compared to other disease aetiologies. Further large-scale studies are required to assess the potential benefits of testosterone therapy.

Keywords:

• Testosterone
• cirrhosis
• non-alcoholic fatty liver disease
• alcohol-related liver disease
• liver transplantation

Disclosure statement of interest

Ross Apostolov, Darren Wong, Elizabeth Low, Karl Vaz, Jessica Spurio, Thomas Worland, Roseanne Kimberley Chan have no conflicts of interest to declare. Paul Gow and Marie Sinclair have received project grants for research from Bayer Pharma AG. Mathis Grossmann has received research funding from Bayer Pharma AG, Otzuka and speaker’s honoraria from Besins Health Care and Novartis.

Data availability statement

The data that support the findings of this study are available from the corresponding author, RA, upon reasonable request.

Additional information

Funding

RA was supported by a National Health and Medical Research Council Postgraduate Fellowship (Grant number 1191028). MG was supported by a National Health and Medical Research Council Project Grant (Grant number 1099173). MS was supported by a National Health and Medical Research Council Early Career Fellowship (Grant number 1138374).