Statins, metformin, and RAS inhibitors did not reduce variceal bleeding risk and mortality in a large, real-life cohort of patients with cirrhosis

Abstract

Background: Previous experimental and clinical studies suggested a beneficial effect of statins, metformin, angiotensin-converting-enzyme inhibitors and angiotensin II receptor blockers (RASi) on portal hypertension. Still, their effects on hard cirrhosis-related clinical endpoints, such as variceal bleeding and bleeding-related mortality, remain to be investigated.

Methods: Thus, we recorded the use of statins, metformin and RASi in a large cohort of cirrhotic patients undergoing endoscopic band ligation (EBL) for primary (PP, n = 440) and secondary bleeding prophylaxis (SP, n = 480) between 01/2000 and 05/2020. Variceal (re-) bleeding and survival rates were compared between patients with vs. without these co-medications.

Results: A total of 920 cirrhotic patients with varices were included. At first EBL, median MELD was 13 and 515 (56%) patients showed ascites. Statins, metformin and RASi were used by 49 (5.3%), 74 (8%), and 91 (9.9%) patients, respectively. MELD and platelet counts were similar in patients with and without the co-medications of interest. Rates of first variceal bleeding and variceal rebleeding at 2 years were 5.2% and 11.7%, respectively. Neither of the co-medications were associated with decreased first bleeding rates (log-rank tests in PP: statins p = 0.813, metformin p = 0.862, RASi p = 0.919) nor rebleeding rates (log-rank tests in SP: statin p = 0.113, metformin p = 0.348, RASi p = 0.273). Similar mortality rates were documented in patients with and without co-medications for PP (log-rank tests: statins p = 0.630, metformin p = 0.591, RASi p = 0.064) and for SP (statins p = 0.720, metformin p = 0.584, RASi p = 0.118).

Conclusion: In clinical practice, variceal bleeding and mortality rates of cirrhotic patients were not reduced by co-medication with statins, metformin or RASi. Nevertheless, we recommend the use of these co-medications by indication, as they may still exert beneficial effects on non-bleeding complications in patients with liver cirrhosis.

Fig 1. Patient flow chart.

Among 1257 patients undergoing endoscopic treatment, 920 patients with esophageal varices were finally included in this study. These patients were divided into groups of primary or secondary prophylaxis and according to the concomitant treatment with statin, metformin and RASi. Abbreviations: N (total numbers), IQR (interquartile range), PHT (portal hypertension), OLT (orthotopic liver transplantation), TIPS (transjugular intrahepatic portosystemic shunt). PVT (portal vein thrombosis) Continuous variables are expressed as median with interquartile range (IQR).

Fig 2. Rates of variceal bleeding.

Kaplan-Meier curves of (re-)bleeding rates for all patients with comedication (statin, metformin or RASi) or without comedication (A-C). Bar chart of first bleeding rates (D) and variceal rebleeding rates (E) within 2 years after first endoscopic band ligation, grouped by intake vs. non-intake of the respective co-medication of interest. n (total numbers), 1Y (Year 1), 2Y (Year 2), 3Y (Year 3), HR (Hazard-Ratio); 95% CI (95% confidence interval); n.s. (not significant); statistical comparisons were performed by log-rank tests.

Fig 3. Mortality rates.

Kaplan-Meier curves showing transplant-free survival for all patients with comedication (statin, metformin or RASi) or non-comedication (A-C). Bar chart of mortality rates in year 2 with comedication or non-comedication for primary prophylaxis and secondary prophylaxis (D,E). n (total numbers), 1Y (Year 1), 2Y (Year 2), 3Y (Year 3), HR (Hazard-Ratio); 95% CI (95% confidence interval); n.s. (not significant); statistical comparisons were performed by log-rank tests.

Fig 4

Kaplan-Meier mortality curves with comedication (statin, metformin or RASi) or non-comedication for primary and secondary prophylaxis (A-F). n (total numbers), 1Y (Year 1), 2Y (Year 2), 3Y (Year 3), HR (Hazard-Ratio); 95% CI (95% confidence interval); n.s. (not significant); statistical comparisons were performed by log-rank tests.

Fig 5. Kaplan-Meier (re-)bleeding curves with NSBB and comedication (statin, metformin or RASi) or non-comedication for primary and secondary prophylaxis.