Real-world safety of terlipressin in elderly patients with liver cirrhosis and esophagogastric variceal bleeding: a focus on hyponatremia and administration methods.

Introduction: Terlipressin is a first-line vasoactive agent for esophageal and gastric variceal bleeding (EGVB) in patients with liver cirrhosis, typically administered alongside proton pump inhibitors as background acid-suppressive therapy. Despite its widespread use, there is limited real-world safety data concerning its effects on the elderly population. This study aims to assess the real-world safety of terlipressin in elderly EGVB patients, with omeprazole administered as standard background therapy, specifically examining the incidence and risk factors associated with hyponatremia and acute kidney injury (AKI). Methods: We conducted a single-center retrospective cohort study involving 90 elderly cirrhotic patients (aged ≥60 years) with EGVB. These patients received a combination of terlipressin and omeprazole at Leshan Hospital of Traditional Chinese Medicine between 1 January 2023, and 31 December 2025. The database was finalized on 31 January 2026, with all patients either discharged or deceased, ensuring no loss to follow-up. We collected data on demographic characteristics, liver function reserve, laboratory indicators, efficacy, and safety indicators. To identify risk factors for hyponatremia, we employed multivariate logistic regression analysis. Additionally, we performed an exploratory comparison of safety differences between two administration methods: intermittent intravenous push and continuous pump infusion. This analysis was Results: The study included 90 patients with an average age of 68.5 ± 6.2 years. Males comprised 70.0% of the cohort, and 38.9% were classified as Child-Pugh class C. The hemostasis success rate within 72 h was 98.9% (89/90), while the rebleeding rate within 5 days was 13.5% (12/89). The in-hospital mortality rate stood at 10.0% (9/90). Hyponatremia occurred in 68.9% (62/90) of patients, with moderate and severe cases (serum sodium <130 mmol/L) accounting for 37.8% (34/90). The incidence of AKI was 17.8% (16/90). Multivariate logistic regression analysis revealed that Child-Pugh class C (OR = 3.12, 95%CI 1.08-9.02, P = 0.03) and reduced baseline serum sodium (per 5 mmol/L decrease, OR = 2.05, 95%CI 1.21-3.47, P = 0.008) were independent risk factors for hyponatremia. Exploratory subgroup analysis showed that the incidence of moderate-to-severe hyponatremia was numerically lower in the continuous pump infusion group than in the intermittent push group (29.2% vs. 51.5%, nominal P = 0.04, uncorrected for multiple comparisons), but there was no significant difference in overall hyponatremia (P = 0.19). Due to the small sample size, this finding requires prospective validation. Conclusion: Terlipressin demonstrates a clear hemostatic effect in elderly patients with liver cirrhosis and EGVB when administered with standard background therapy including omeprazole. However, it is associated with a relatively high incidence of hyponatremia. Patients classified as Child-Pugh class C or those with low baseline serum sodium levels require enhanced monitoring of serum sodium. The impact of administration method on hyponatremia requires further investigation.