The early distinction between prerenal azotemia, characterized by an avid proximal tubular sodium reabsorption, and ATN, in which proximal tubule function is depressed, remains an important but difficult clinical task. Indices of acute renal failure based on urinary sodium excretion may be helpful but have several limitations, among which is the use of diuretics. The effectiveness of the fractional excretion of uric acid (FEUA) and that of endogenous lithium (FELi) in the diagnosis of acute renal failure has been evaluated in an unselected group of 46 patients, 28 with prerenal azotemia and 18 with ATN. In the entire group, FELi concurred with the clinical diagnosis in 78% of the patients, whereas the fractional excretion of sodium (FENa) and FEUA were in agreement in only 63 and 50%, respectively. FELi was more sensitive to identify hemodynamic renal failure, because 93% of prerenal failure patients had a low FELi, contrasting with a low FEUA in only 68% and a low FENa in 75%. The major reason for the discrepancy between FENa and FELi was the administration of diuretics. In both acute renal failure groups, FENa was higher in the subgroups receiving diuretics. In contrast, diuretic therapy had no effect on FELi in either group. These results suggest that FELi is more accurate than either FENa or FEUA for distinguishing prerenal azotemia from ATN. The superiority of FELi appears especially relevant in patients treated with the usual diuretics.