Background and aim: Accurate assessment of cirrhotic patient's prognosis is essential for decisions regarding the course of treatment. Therefore we aimed to confirm and quantify the predictive value of serum cholesterol and serum triglycerides in liver cirrhosis patients.
Material and methods: We performed a retrospective observational cohort study on consecutive patients with liver cirrhosis (n = 191). Relevant clinical and laboratory variables were obtained from patients' charts and patients were followed for two months. Mortality was the main outcome.
Results: Thirty-eight patients died in the follow-up period. Significant difference was observed in the level of total serum cholesterol between surviving and deceased patients (2.27 ± 1.02 mmol/L vs. 2.97 ± 1.00 mmol/L, P < 0.0001 respectively). Cholesterol was confirmed as a significant predictor of mortality in univariate logistic regression analysis, and independent predictor beside bilirubin, creatinine and MELD score in multivariate logistic regression analysis. Addition of serum cholesterol level to a prognostic model based on total bilirubin, creatinine and INR increased its accuracy by 4%. Adding cholesterol to the MELD score improved prediction accuracy by 3%. There was no significant difference in serum levels of triglycerides between surviving and deceased patients.
Conclusion: Serum cholesterol is a routinely measured parameter, which has independent prognostic value in patients with liver cirrhosis.