Rapid estimation of creatinine clearances in patients with liver dysfunction

Different methods are available for rapidassessment of renal function using the patient's serumcreatinine concentration and body weight withoutobtaining urine collection over 24 hr. However, thereliability of these methods in patients with liverdiseases has not been established. The purpose of thisstudy was to determine the accuracy and precision of theestimated creatinine clearances obtained by the methods of Cockcroft-Gault, Jelliffe, Mawer, andSiersbaek-Nielsen in patients with liver diseases whohave different degrees of renal function. Creatinineclearances obtained from 24-hr urine collection wereused as the standard. The different methods forrapid renal function estimation had limited accuracy andreliability in patients with severe liver dysfunction(Child-Pugh class C) and also in those with creatinine clearances of less than 60 ml/min. Creatinineclearances were overestimated by about 40-100%. Usinglean body weights, instead of total body weights,reduced the prediction errors. In patients with mild liver dysfunction (Child-Pugh class A), allfour estimation methods provided reasonable estimationof the creatinine clearances.