Serum cystatin C as a predictor of 90-day mortality among patients admitted with complications of cirrhosis

Background and Aim: Cystatin C (Cys) is not affected by age, sex, and muscle mass. We evaluated to compare the predictive performance of serum Cys level and model for end-stage liver disease (MELD) score and developed a new model to predict 90-day mortality among patients admitted with cirrhosis complications. Methods: A prospective cohort study was performed from December 2018 to December 2019. All cirrhotic patients admitted with acute decompensated liver cirrhosis or acute on chronic liver failure had laboratory values measured within 48 h of admission. Results: A cohort of 225 patients with cirrhosis was admitted during the study period. Sixty-five patients were eligible for analysis. Twenty-seven of these patients (41.4%) died within 90 days of follow-up. The median of MELD score was 20.5 (15, 24). Serum Cys level of >1.45 mg/L had the highest 90-day mortality prediction with the sensitivity and specificity of 66.7% and 68.4%, respectively. Cys and MELD scores were predictive of 90-day mortality: Cys hazard ratio (HR) = 2.04 (95% CI 1.01-4.14, P = 0.048); MELD score HR = 1.01 (95% CI 0.51-2.01, P = 0.970). C-statistic of Cys, MELD score, model for end-stage liver disease-cystatin C (MELD-Cys) score, combined Cys with MELD-Cys score to predict 90-day mortality were 0.67, 0.58, 0.58, and 0.63, respectively. Adding Cys to the MELD score did not improve the predictive of 90-day mortality. Conclusion: Serum Cys is superior to MELD score, and the new MELD-Cys model is comparable to the MELD score in predicting mortality among patients with cirrhosis admitted with complications.