Testosterone Replacement Reduces Morbidity and Mortality for Most Patients With Cirrhosis.

Testosterone replacement was associated with lower all-cause mortality in men with cirrhosis and hypogonadism.

• Subdistribution hazard ratio (sHR) for mortality was 0.92 (95% CI, 0.85–0.99).
• A total of 282 patients (7.4%) with testicular hypofunction and cirrhosis received testosterone replacement after diagnosis.
• Analysis used inverse probability of treatment weighting to balance treated and untreated groups.
• Outcomes were evaluated using an intention-to-treat design.

Testosterone replacement was associated with a lower risk of new decompensation events overall.

• sHR for new decompensation events was 0.92 (95% CI, 0.86–0.99).
• The association was particularly strong for ascites requiring paracentesis (sHR, 0.82; 95% CI, 0.76–0.89).
• Variceal hemorrhage risk was also substantially reduced (sHR, 0.67; 95% CI, 0.54–0.85).
• The effect on hepatic encephalopathy requiring hospitalization was less pronounced (sHR, 0.92; 95% CI, 0.84–1.01), with a confidence interval crossing 1.

Testosterone replacement was not associated with a reduced risk of fractures or an increased risk of hepatocellular carcinoma.

• sHR for fractures was 0.99 (95% CI, 0.91–1.08), indicating no meaningful effect.
• sHR for hepatocellular carcinoma was 1.09 (95% CI, 0.91–1.3), indicating no significant increased risk.
• Both confidence intervals crossed 1, suggesting no statistically significant association for either outcome.

Actual testosterone therapy exposure differed substantially between patients started on testosterone and those who were not.

• Patients started on testosterone spent 28.6% of patient-days on therapy.
• Patients not started on testosterone spent only 0.5% of patient-days on therapy.
• The difference was statistically significant (P < .0001).

The study was conducted as an emulated clinical trial using a new user design with nationally representative Medicare data.

• Medicare data from 2008 to 2020 were used.
• Patients were required to have been diagnosed with hypogonadism at the same time as their diagnosis of cirrhosis (new user design).
• The total cohort included men with both testicular hypofunction and cirrhosis, of whom 282 (7.4%) received testosterone replacement.
• There was substantial heterogeneity of treatment effect across baseline subgroups.