Limitations of Wright formula estimates of renal function

Abstract

It was with interest that we read the publication by Marx et al. [1.Marx G.M. Blake G.M. Steer C.B. et al.Evaluation of the Cockcroft-Gault, Jelliffe and Wright formulae in estimating renal function in elderly cancer patients.Ann Oncol. 2004; 15: 291-295Abstract Full Text Full Text PDF PubMed Scopus (123) Google Scholar] who examined a number of formulae to estimate renal function in cancer patients over 70 years of age. They concluded that the Wright formula would provide an adequate estimation of glomerular filtration rate (GFR) in patients with a GFR within the normal range (50–120 ml/min). However, the key practical question relating to this work was not addressed, that is, the performance of the formulae in accurately predicting renal function for all patients across the full range of renal function, and particularly those with poor renal function (GFR <50 ml/min). This is an especially relevant issue for elderly patients, who were the patient group in this study. It would appear from the data that there were 40 patients with a GFR <50 ml/min; however, no details of the bias and precision of the Wright formula was presented for this group. Visual inspection of the data presented (Marx et al. [1.Marx G.M. Blake G.M. Steer C.B. et al.Evaluation of the Cockcroft-Gault, Jelliffe and Wright formulae in estimating renal function in elderly cancer patients.Ann Oncol. 2004; 15: 291-295Abstract Full Text Full Text PDF PubMed Scopus (123) Google Scholar], figure 1A) would indicate that the Wright formula significantly overestimates GFR in patients with poor renal function. The authors make no comment on the utility of the formula for this group of patients. It is worth noting also, that the original Wright paper did not provide an analysis for low levels of renal function [2.Wright J.G. Boddy A.V. Highley M. et al.Estimation of glomerular filtration rate in cancer patients.Br J Cancer. 2001; 84: 452-459Crossref PubMed Scopus (156) Google Scholar]. In their paper, Marx et al. comment that our group have also shown the Wright formula to be superior over the Cockroft–Gault and Jelliffe formulae [3.Poole S.G. Dooley M.J. Rischin D. A comparison of beside renal function estimates and measured glomerular filtration rate (Tc99m DTPA clearance) in cancer patients.Ann Oncol. 2002; 13: 949-955Abstract Full Text Full Text PDF PubMed Scopus (62) Google Scholar]. In our study this was correct when comparing the various formulae over all ranges of GFR; however, we actually showed that the Wright formula did have a significant positive bias for low GFR (i.e. overestimates) and a significant negative bias for high GFR (i.e. underestimates). This led us to conclude in our paper that the Wright formula provides a biased and imprecise estimate of GFR. This interpretation is further supported in that we have recently analysed data on 525 patients (as yet unpublished) who have had GFR determined by Tc99m DTPA clearance at our centre. In this group there were 41 patients >70 years of age with a GFR <50 ml/min. The mean percentage error of the Wright formula was 26.5% (95% confidence interval 13.1% to 39.8%). Consequently, we cannot recommend the use of the Wright formula. From a clinical perspective, a tool is needed that provides a reliable estimate across the full range of renal function. Our data, and possibly that of Marx et al. [1.Marx G.M. Blake G.M. Steer C.B. et al.Evaluation of the Cockcroft-Gault, Jelliffe and Wright formulae in estimating renal function in elderly cancer patients.Ann Oncol. 2004; 15: 291-295Abstract Full Text Full Text PDF PubMed Scopus (123) Google Scholar], indicate that the Wright formula has substantially limited applicability. Further research is required to develop reliable methods of bedside estimates of GFR, as there is no precise formula available. Recently, the Modification of Diet in Renal Disease formula has been advocated, but this too requires validation in cancer patients with poor renal function [4.Levey A.S. Bosch J.P. Breyer L.J. et al.A more accurate method to estimate glomerular filtration rate from serum creatinine: A new prediction equation.Ann Intern Med. 1999; 130: 461-470Crossref PubMed Scopus (13045) Google Scholar, 5.Murray P.T. Ratian M.J. Estimation of the glomerular filtration rate in cancer patients: a new formula for new drugs.J Clin Oncol. 2003; 21: 2633-2635Crossref PubMed Scopus (26) Google Scholar].