Lithium clearance in the evaluation of segmental renal tubular reabsorption of sodium and water in diabetes mellitus.

Lithium is the best available marker of proximal tubular reabsorption of fluid. The first part of the present thesis reviews the background for the use of the lithium clearance (CLi) method. Micropuncture studies on proximal reabsorption of lithium, showed that CLi is a reasonably correct measure of end-proximal fluid delivery rate, even during osmotic diuresis. During severe salt restriction, distal reabsorption of lithium renders the CLi method inappropriate in animals, but this problem does probably not occur in humans. The major current issue is whether a quantitatively significant reabsorption of lithium occurs in the loop of Henle. Available evidence is in accord with the interpretation that it does not occur. The interpretation of results form CLi studies depends to a surprising degree on the investigators beliefs about renal physiology. In the evaluation of proximal tubular function, the relevant parameter is the absolute proximal reabsorption rate of fluid and sodium. In the evaluation of integrated distal tubular reabsorption of sodium, the relevant parameter is the fractional distal reabsorption rate of sodium. The fractional CLi does not give meaningful information, and calculated absolute distal reabsorption rate of sodium is inherently not suited to detect modest changes in distal reabsorption leading to large changes in sodium excretion. Results from the use of the CLi method in relation to diabetes are reviewed in the second section. Even in IDDM patients with early diabetic nephropathy, the proximal reabsorption rate is elevated, resulting in a normal CLi despite glomerular hyperfiltration. Overnight euglycemia did not change GFR in IDDM patients, but during maintained euglycemia, GFR was normalized. A few hours of hyperglycemia prevented the decline in GFR, whereas CLi was unchanged. Thus hyperglycemia produced changes in renal function similar to those observed previously, but the time-course of the effect of euglycemia on kidney function is delayed. Plasma levels of atrial natriuretic peptide, renin and glucagon were not importantly affected by plasma glucose. In NIDDM patients CLi was normal, despite slight hyperfiltration, although this observation must be confirmed in a study with larger sample size. Prompted by the clinical observation of a marked decline in the GFR induced by carbonic anhydrase inhibitors, we studied the renal effects of acetazolamide in a controlled study.(ABSTRACT TRUNCATED AT 400 WORDS)