Lithium and free water clearance (CLi and CH2O) were measured simultaneously in 5 patients with Bartter's syndrome (BS), and the results were compared with values obtained in a large group of healthy control subjects. In line with the literature, fractional diluting segment reabsorption estimated from CH2O appeared to be subnormal in BS (82.4 +/- 4.7 versus 89.4 +/- 4.4% in controls, p < 0.01). Fractional delivery to this segment, expressed as the maximum urine flow during water diuresis (Vmax/glomerular filtration rate, GFR), also appeared to be reduced (11.3 +/- 1.6 versus 13.8 +/- 2.5%, p < 0.05), suggesting compensatory increased proximal reabsorption. In disagreement with the latter, CLi/GFR, alleged to be a measure of fractional sodium and water output from the proximal tubules, was increased (36.2 +/- 1.8 versus 30.7 +/- 5.1% in controls, p < 0.01), suggesting that suppressed proximal reabsorption could be the primary reabsorption defect in BS. During chronic converting enzyme inhibition with enalapril, given to treat hypokalemia (3 patients), the concordance between the two methods was restored because CLi/GFR fell by approximately 50%, whereas VMax/GFR did not change; the alleged diluting segment reabsorption defect remained present. These contradictory results lead to the conclusion that one of the two methods reflected the pattern of tubular sodium handling in BS incorrectly. However, the present study does not unequivocally indicate which method yielded erroneous results in this condition.