Background: The accuracy of prediction equations has not been validated in adult patients with chronic kidney disease (CKD) stages 4–5 in extreme situations of nutritional status and age. Objective and Methods: The significance of nutritional status, calculated with the creatinine production (CP) formula, and age (≤64 years and >64 years) in the application of different prediction equations – modification of diet in renal disease (MDRD), simplified MDRD (sMDRD), Cockcroft-Gault (CG) – and the mean of urea and creatinine clearance (Cr-Ur) compared with the isotopic glomerular filtration rate (GFR) estimation calculated by 51Cr-EDTA was studied in 87 Caucasian adults with CKD stages 4–5 (GFR: 30–8 ml/min/1.73 m<sup>2</sup>). The Bland-Altman method and Lin’s concordance coefficient (Rc) were used to study accuracy (bias) and precision. Results: The GFR calculated with <sup>51</sup>Cr-EDTA in the study group was 22.2 ± 6.9 ml/min/1.73 m<sup>2</sup> (range: 8–30). CG and sMDRD were the best prediction equations with bias of –1.1 and –3.8 ml/min/1.73 m<sup>2</sup> and Rc of 0.52–0.50. In this situation, the mean Cr-Ur proved the most inaccurate equation compared with the isotopic technique with bias of –5.4 ml/min/1.73 m<sup>2</sup> and Rc of 0.32. In the analysis of patients with higher CP (> 0.90; n = 44), CG and sMDRD obtained the best bias of 1.2 and –2.7 ml/min/1.73 m<sup>2</sup> and Rc of 0.54–0.53. In patients aged ≤64 (n = 44), these equations obtained a bias of 1.1 and –3.6 ml/min/1.73 m<sup>2</sup> and Rc 0.50–0.49. Both in lower CP (≤0.90; n = 43) and older age (>64 years; n = 43), all the equations underestimated the value obtained with isotopic GFR. In these situations, the results obtained with CG had a bias of –2.2 and –3.6 ml/min/1.73 m<sup>2</sup> (Rc 0.29–0.56) and with sMDRD –4.0 and –4.1 ml/min/1.73 m<sup>2</sup> (Rc 0.39–0.51). In these circumstances, Cr-Ur was the most inaccurate equation, obtaining a bias of –10.1 and –13.2 ml/min/1.73 m<sup>2</sup> (Rc 0.14–0.16). Conclusions: In the group with higher CP and age ≤64 years, results of the presented data yielded no evidence for superiority of the MDRD equation over CG formula in patients with advanced renal failure. On the basis of our results, we do not recommend the use of the Cr-Ur adjusted to 1.73 m<sup>2</sup> of body surface area, which was the most imprecise equation. Application of all the equations proved inaccurate in lower CP patients with or without advanced age, implying the premature start of substitution renal treatment. In these circumstances, ambulatory GFR determination by isotopic techniques would be indicated.