Does lithium clearance reflect distal delivery in humans? Analysis with furosemide infusion

Fractional excretion of lithium (FELi+) has been proposed as an index of fluid delivery to the distal nephron. The increase in FELi+ after the "loop diuretic" furosemide indicates that this postulate may not be valid unless furosemide acts in the proximal tubules. We studied the effect of furosemide (40 mg iv as bolus, followed by 20 mg/h infusion for 90 min) in eight healthy male subjects during maximal water diuresis. Special care was taken to exactly replace urinary losses. Furosemide greatly increased fractional excretion of sodium, from 1.3 +/- 0.4 to 27.8 +/- 3.9%, and water, from 14.2 +/- 1.7 to 38.2 +/- 3.7% (P less than 0.01). There was a disproportionately large increase in FELi+ from 30.3 +/- 3.0 to 53.7 +/- 2.9% (P less than 0.01), whereas fractional excretion of some other alleged proximal markers increased to a lesser extent. Lysine vasopressin, infused at the end of the experiment (n = 7), caused only a small increase in urine osmolality from 225 +/- 17 to 241 +/- 17 mosmol/kg (P less than 0.01), indicating that medullary hyperosmolality had been largely abolished. The most likely explanation of these results is that furosemide has a moderate action in the proximal tubules, and at the same time inhibits preexistent lithium absorption in Henle's loop. In addition, the large difference between FELi+ and maximal urine flow remaining after furosemide suggests that, despite the decreased medullary osmotic gradient, water backdiffusion is unaltered by furosemide or that lithium concentration in the proximal tubule is changed by furosemide.