Background and objective: Accurate quantification of renal function is important for diagnosing and monitoring progression of renal diseases and for calculating adequate doses of drugs that are excreted by the kidneys. Gold-standard procedures are too complex for routine clinical use. At the moment there are several formulae to choose from, all said to estimate renal function precisely enough for clinical purposes. It was the aim of this study to compare the accuracy of several of these in clinical routine.
Patients and methods: The results of inulin clearance were compared with those calculated by the Cockcroft-Gault formula (CGF), abbreviated diet modification of renal disease (MDRD) formula, the Mayo formula and the cystatin C-based formula as proposed by Larsson et al. Included were 189 in-patients (aged 20-87, 40% of them women, range of inulin clearance 8-244 ml/min/1,73m). In addition, inulin clearance was compared with creatinine clearance in 142 patients (aged 20-87 years, 42% women. Inulin clearance 13-244 ml/min/1,73m). Bland-Altman diagrams were drawn and mean bias and standard deviation of the formulae were compared with inulin clearance, as were sensitivity and specifity for diagnosing reduced renal function.
Results: All formulae underestimated glomerular filtration rate (GFR), with CGF and MDRD formulas giving the best results. These formulae had a mean bias of -16.2 (SD 24.8) and -18.2 (SD 25.6) ml/min/1,73m (2) , respectively. All creatinine-based formulae showed a high sensitivity and specifity for diagnosing a GFR below 60 ml/min/1,73m (2).
Conclusion: None of the estimating formulae can replace inulin clearance with adequate accuracy. In our patients the cystatin C formula of Larsson et al showed no advantage. But the MDRD formula, which can be calculated without knowing body weight, is as accurate and precise as CGF.