Cytokeratin-18 (CK-18) is a marker of hepatic cell death. Serum CK-18 could serve as a prognostic marker for patients with advanced liver disease (ALD). This meta-analysis aims to explore the association between total CK-18 (M65) and caspase-cleaved CK-18 (M30) levels with the prognosis of ALD patients. Relevant longitudinal observational studies were identified through comprehensive searches of the Medline, Web of Science, and Embase databases. A random-effects model was utilized to synthesize the findings, accommodating heterogeneity among studies. The analysis included 14 datasets from 11 studies. Elevated serum CK-18 levels at admission were linked to a higher risk of death or liver transplantation during follow-up. This association was consistent for both M65 (risk ratio [RR] 1.99, 95% confidence interval [CI] 1.65 to 2.40, p < 0.001; I2 = 43%) and M30 (RR 1.94, 95% CI 1.57 to 2.40, p < 0.001; I2 = 46%). Subgroup analysis revealed that the relationship between serum M65 levels and adverse outcomes was attenuated in studies using multivariate analysis compared to those using univariate analysis (RR 1.78 vs. 2.80, p for subgroup difference = 0.03). Further subgroup analyses indicated that the prognostic significance of CK-18 for ALD patients was not significantly influenced by study design, methods of determining CK-18 cutoff values, or follow-up durations. Elevated serum CK-18 levels at admission indicate a poor prognosis in patients with ALD. This finding holds for both M65 and M30.