Aim: The assessment of renal function is of vital importance in management of patients with cirrhosis. While serum creatinine (Cr) is routinely used for this purpose, Cr-based estimated glomerular filtration rate (eGFR) does not reflect true renal function because of muscle wasting and impaired liver function. By contrast, cystatin C (CysC) is unrelated to muscle volume and liver function. In this study, we examined whether CysC-based GFR estimation is beneficial in assessment of renal function in patients with cirrhosis.
Methods: First, we assessed the performance of GFR-predicting equations based on serum Cr or CysC in 14 patients with cirrhosis, by comparison with inulin clearance as a gold standard of GFR (measured GFR [mGFR]). Next, in 49 patients with cirrhosis, we examined serum Cr and CysC at baseline, and examined which GFR-predicting equations were more suitable for predicting the outcome.
Results: In the first experiment, mGFR was 54.3 ± 23.0 mL/min, and CysC-based GFR-estimating equations had better performances compared with Cr-based equations in terms of bias, precision and accuracy. Cr-based estimated GFR (eGFRcreat) was significantly different from mGFR (P < 0.05). In the follow-up study of 49 patients (observational period, 30.7 ± 32.0 months), multivariate analysis demonstrated that CysC-based estimated GFR (eGFRcys), along with albumin, Child-Pugh grade and presence of hepatocellular carcinoma, was independently associated with overall survival (odds ratio, 4.19; 95% confidence interval, 1.44-12.2, P = 0.009).
Conclusion: These results suggest that eGFRcys could estimate renal function and predict outcome more accurately compared with Cr-based eGFR in cirrhotic patients.
Keywords: creatinine; cystatin C; glomerular filtration rate; outcome.
© 2015 The Japan Society of Hepatology.