Glomerular filtration rate is an independent factor of mortality in patients with decompensated cirrhosis.

Aim: Although serum creatinine is included in the Model for End-Stage Liver Disease (MELD) score, it is an inaccurate marker of renal function, namely, of glomerular filtration rate ("true" GFR) in patients with decompensated cirrhosis. Our aim was to investigate the impact of MELD score and "true" GFR as determinants of survival in patients with decompensated cirrhosis.

Methods: We included all consecutive patients with decompensated cirrhosis who were admitted to our department. Renal function was assessed by creatinine- and cystatin-based estimated GFR and "true" GFR using (51) Cr-ethylenediaminetetraacetic acid. The independent factors associated with survival were evaluated. The discriminative ability of the prognostic scores (MELD and modifications of MELD score) were evaluated by using the area under the receiver-operator curve (AUC).

Results: One hundred and ten consecutive patients (77 men, aged 56 ± 12 years); at the end of follow up (8 months; range, 6-18), 92 patients (84%) were alive and 18 (16%) had died. In multivariate analysis, serum bilirubin (hazard ratio [HR], 1.15; 95% confidence interval [CI], 1.05-1.26; P = 0.020) and "true" GFR (HR, 0.96; 95% CI, 0.93-0.98; P = 0.003) were the only independent factors significantly associated with the outcome. The derived new prognostic model had high discriminative ability (AUC, 0.90), which was confirmed in the validation sample of 77 patients.

Conclusion: In our cohort of patients with decompensated cirrhosis, "true" GFR and bilirubin were the independent factors of the outcome.