Glucagon-like Peptide-1 Receptor Agonist Use in Patients With Liver Cirrhosis and Type 2 Diabetes

Background and Aims Liver cirrhosis is often associated with type 2 diabetes (T2D), but research on treatment of T2D in cirrhotic patients is scarce. We investigated the long-term outcomes of glucagon-like peptide-1 receptor agonists (GLP-1 RAs) in patients with T2D and cirrhosis. Methods Using propensity score matching, we selected 467 matched pairs of GLP-1 RA users and nonusers from the National Health Insurance Research Database of Taiwan from January 1, 2008, to December 31, 2019. Multivariable-adjusted Cox proportional hazards models were used to compare the outcomes between GLP-1 RA users and nonusers. Results The mean follow-up time was 3.28 and 3.06 years for GLP-1 RA users and nonusers, respectively. The rates of death were 27.46 and 55.90 per 1000 person-years for GLP-1 RA users and nonusers, respectively. The multivariable-adjusted models showed that GLP-1 RA users had lower risks of mortality (adjusted hazard ratio [aHR], 0.47; 95% confidence interval [CI], 0.32–0.69), cardiovascular events (aHR, 0.6; 95% CI, 0.41–0.87), decompensated cirrhosis (aHR, 0.7; 95% CI, 0.49–0.99), hepatic encephalopathy (aHR, 0.59; 95% CI, 0.36–0.97), and liver failure (aHR, 0.54; 95% CI, 0.34–0.85) than nonusers. A longer cumulative duration of GLP-1 RA use had a lower risk of these outcomes than GLP-1 RA nonuse. Conclusions This population-based cohort study showed that GLP-1 RA users exhibited a significantly lower risk of death, cardiovascular events, decompensated cirrhosis, hepatic encephalopathy, and liver failure in patients with T2D and compensated liver cirrhosis. Additional studies are needed to confirm our results. Liver cirrhosis is often associated with type 2 diabetes (T2D), but research on treatment of T2D in cirrhotic patients is scarce. We investigated the long-term outcomes of glucagon-like peptide-1 receptor agonists (GLP-1 RAs) in patients with T2D and cirrhosis. Using propensity score matching, we selected 467 matched pairs of GLP-1 RA users and nonusers from the National Health Insurance Research Database of Taiwan from January 1, 2008, to December 31, 2019. Multivariable-adjusted Cox proportional hazards models were used to compare the outcomes between GLP-1 RA users and nonusers. The mean follow-up time was 3.28 and 3.06 years for GLP-1 RA users and nonusers, respectively. The rates of death were 27.46 and 55.90 per 1000 person-years for GLP-1 RA users and nonusers, respectively. The multivariable-adjusted models showed that GLP-1 RA users had lower risks of mortality (adjusted hazard ratio [aHR], 0.47; 95% confidence interval [CI], 0.32–0.69), cardiovascular events (aHR, 0.6; 95% CI, 0.41–0.87), decompensated cirrhosis (aHR, 0.7; 95% CI, 0.49–0.99), hepatic encephalopathy (aHR, 0.59; 95% CI, 0.36–0.97), and liver failure (aHR, 0.54; 95% CI, 0.34–0.85) than nonusers. A longer cumulative duration of GLP-1 RA use had a lower risk of these outcomes than GLP-1 RA nonuse. This population-based cohort study showed that GLP-1 RA users exhibited a significantly lower risk of death, cardiovascular events, decompensated cirrhosis, hepatic encephalopathy, and liver failure in patients with T2D and compensated liver cirrhosis. Additional studies are needed to confirm our results.