Chronic kidney disease (CKD) is a world-wide public health problem, with adverse outcomes of kidney failure, cardiovascular disease, and premature death. The National Kidney Foundation, through its Kidney Disease Quality Outcome Initiative (K/DOQI) and other National institutions, recommend glomerular filtration rate (GFR) estimates for the definition, classification, screening, and monitoring of CKD. Prediction equations based on serum creatinine values were chosen both for adults (Cockcroft-Gault [C-G] and Modification of Diet in Renal Disease [MDRD] study equations) and for children (Schwartz and Counahan-Barratt equations). This review aims to evaluate from recent literature the clinical efficiency and relevance of these equations in terms of bias, precision, and reproducibility in different specific indications (eg, screening CKD, assessment of disease progression, or therapy efficacy) in different populations. Because these prediction equations based on serum creatinine have limitations, especially in the normal or near-normal GFR range, kidney transplant recipients, and pediatric populations, other prediction equations based on serum cystatin C value were also considered as possibly more sensitive GFR surrogate markers. Recent guidelines state that the cystatin C-based prediction equation cannot be recommended for use in clinical practice. With prediction equations based on serum creatinine, the National Kidney Disease Education Program (NKDEP) recommendations are to report a numerical estimate in round numbers only for GFR values <60 mL/min per 1.73 m(2). The MDRD equation generally outperforms the C-G equation but may still have a high level of bias, depending on creatinine assay calibration, and low precision with, at best, approximately 80% of estimated GFR in the "accuracy range" of 70-130% of the measured GFR value, even in patients with known CKD. According to Kidney Disease Improving Global Outcomes (KDIGO) recommendations, many indications remain for GFR measurements using a clearance method. In that context, it should be recalled that radiolabeled-tracer plasma or urinary clearance methods, are safe, simple, accurate and reproducible.