Background: Discrepant results have been published regarding the suitability of creatinine clearance (C(Cr)) as a measure of glomerular filtration rate (GFR) in cirrhotic patients with normal renal function.
Subjects and methods: In this study we evaluated the accuracy and precision of measured and calculated C(Cr) as indexes of GFR by comparing their values to those of inulin clearance (C(In)) in 10 healthy subjects and 20 patients with either Child's class A or Child's class C liver cirrhosis.
Results: The accuracy and precision of GFR estimates obtained by measuring C(Cr) were good in all three study groups. The mean values of the C(Cr)/C(In) ratio were 1.05, 1.03 and 1.04, respectively, and the corresponding coefficients of variations were 2.9, 2.9 and 3.8%. A close correlation between C(Cr) and C(In) was also found in each study group (r = 0.98, 0.99 and 0.97, respectively, with p < 0.001 in each case). C(Cr) calculated from serum creatinine by means of the Cockcroft-Gault formula (predicted GFR) proved to be a suitable measure of GFR in normal subjects and patients with Child's class A cirrhosis: the predicted-to-true GFR ratios were 0.93 and 0.94, respectively, CV was 12% in both cases. Moreover, a significant correlation between predicted and true GFR was observed in both groups (r = 0.73, p < 0.02 and r = 0.69, p < 0.025, respectively). On the contrary, in Child's class C cirrhotics, calculated C(Cr) significantly overestimated GFR (predicted-to-true GFR ratio 1.23, CV 20%) and no significant correlation was found between predicted and true GFR (r = 0.58, p > 0.05).
Conclusion: In conclusion, this study shows that measured C(Cr) is a reliable index of GFR in cirrhotic patients, irrespective of the degree of liver dysfunction. Calculated C(Cr) is still an adequate marker of GFR in patients with compensated liver cirrhosis, whereas it overestimates GFR in patients with decompensated cirrhosis. A lower muscle mass, a reduced ability to convert creatine to creatinine, and the presence of ascites are most likely responsible for the overestimation of GFR by the Cockcroft-Gault formula in the latter patients.