Urinary ammonium excretion is the best parameter to quantify net acid excretion by the kidney. As measurement of ammonium excretion is not routinely available, clinicians use formulas to estimate NH(4)+ excretion. However, the measurement of urinary NH(4)+ concentration can be performed by an automatically method, which is suitable for clinical practice. The aim of this study is to evaluate the validity of the enzymatic method of ammonium determination with Ammonia SL Elitech Diagnostics reagent on an Olympus AU 2700 analyzer, which use 1/100 diluted urine samples. A clinico-biological study allowed us to compare measurements obtained during a 30 months' period with the above enzymatic method with results obtained by a formula of calculation. Variations coefficients (CV%) of repeatability were less than 2.4% and, those of reproducibility tests less than 2,6%. Linearity was verified from 0.62 mmol/L to 158 mmol/L. Analytical sensitivity was 0.52 mmol/L and the correlation obtained with the assay used to date in the laboratory was excellent (y = 1.11 x - 1.72 ; r = 0.98). There is a significant positive correlation between measured concentrations obtained with this enzymatic method and urinary ammonium concentration estimates using the modified urine osmolal gap in two groups of patients, with and without mild chronic renal failure. As urine ammonium estimation is not reliable for detecting small changes in ammonium excretion, it must be absolutely measured when renal functional tests are performed. The assay described in this paper is simple, automatic and offers for the clinician accurate matter for the measurement of NH(4)+ excretion.