Inaccuracy of the Schwartz formula in estimating glomerular filtration rate in Nigerian children.

ABSTRACT

The accurate estimation of renal function is of vital importance in the management of a child with renal disease. Given the well known difficulties of obtaining an accurately timed 24-hour urine sample from children and lacking the resources for estimating glomerular filtration rate (GFR) using radionuclide methods, the clinician practising in a developing country often has to resort to the use of height/plasma creatinine formulae when a rapid assessment of GFR is needed. In order to assess the accuracy of one of the better known formulae (Schwartz formula) in predicting GFR, 34 children with the nephrotic syndrome and 30 apparently healthy children with no evidence of renal disease were studied at the University College Hospital, Ibadan. Two methods, Altman-Bland analysis and correlation coefficients, were used to assess agreement between measured GFR (by endogenous creatinine clearance) and GFR estimated by formula. The results show that the height/plasma creatinine formula of Schwartz et al. is a poor predictor of GFR as measured by endogenous creatinine clearance in Nigerian children and that it overestimated GFR in over two-thirds of the children. These observations may be due to differences in the constant, k, in the formula, which was found to vary widely in this study with a mean value of 0.45 compared with 0.55 in the formula. It is suggested that height/creatinine formulae for predicting GFR be tested and validated for accuracy in a given environment before routine use in clinical settings.