Diuretic effect of frusemide in patients with nephrotic syndrome: is it potentiated by intravenous albumin?

BMJ 1995; 310 doi: (Published 21 January 1995) Cite this as: BMJ 1995;310:162
  1. Fehmi Akcicek, assistant professora,
  2. Turkay Yalniz, medical studenta,
  3. Ali Basci, head of renal departmenta,
  4. Ercan Ok, postdoctoral fellow in nephrologya,
  5. Evert J Dorhout Mees, professor of nephrologya
  1. a Ege University School of Medicine, Department of Internal Medicine, Division of Nephrology, Bornova 35100, Izmir, Turkey
  1. Correspondence to: Dr Akcicek.
  • Accepted 25 November 1994

We investigated the claim that infusion of albumin potentiates the diuretic effect of frusemide in patients with the nephrotic syndrome.1

Patients and methods

We selected 12 inpatients with the nephrotic syndrome according to two criteria: (a) failure to lose weight after bed rest and a 40 mmol sodium diet and (b) presence of minimal lesions or a well preserved glomerular histology on renal biopsy. We gave each patient three treatments randomly from 8 am to noon, at intervals of at least two days—infusion of albumin 0.5 g/kg given as a 20% solution for four hours; infusion of frusemide 60 mg bolus followed by 40 mg/h for four hours; and a combined treatment of both types of infusion. We collected urine on the treatment days from 8 am to 2 pm and from 2 pm through to 8 am the next day and on control days from 8 am to 8 am the next day.

Four patients were excluded from the study (one became temporarily anuric, two lost their oedema before completion, and one failed to comply). Histological examination in the remaining eight patients showed minimal lesions (six patients), amyloidosis (one), and membranoproliferative nephritis (one). Serum creatinine concentrations ranged from 106.1 to 212.2 μmol/l, plasma albumin concentrations were 11 to 22 g/l, and blood pressure ranged from 100/65 to 140/90 mm Hg. The treatment sequences (each of them in two patients) were albumin, frusemide, albumin and frusemide; albumin, albumin and frusemide, frusemide; frusemide, albumin and frusemide, albumin; albumin and frusemide, albumin, frusemide. Care was taken that grossly apparent clinical oedema was present at the beginning of each treatment.


During the infusion of albumin, plasma albumin concentrations increased from 17.3 (SD 1.1) to 23.6 (2.3) g/l while the packed cell volume decreased from 0.33 (0.01) to 0.27 (0.01), resulting in a calculated increase in plasma volume of 30%. During the combined infusion of albumin and frusemide the corresponding values were 17.0 (1.2) to 23.4 (2.4) g/l and 0.33 (0.01) to 0.28 (0.01) respectively.

The table shows the mean changes in volume of urine and the mean excretions of sodium and potassium during the three different treatments. Albumin alone caused a small but negligible increase in volume of urine and excretion of sodium in all patients. In contrast, frusemide induced more than a 10-fold increase in volume of urine and 60-fold increase in excretion of sodium. The effect of combined albumin and frusemide was similar to that of frusemide alone. Excretion of potassium increased to the same degree (3–5 times control) after frusemide and combined albumin and frusemide. During the 18 hours immediately after infusion of albumin, volume of urine and excretion of sodium remained slightly raised compared with previous control days. During the control days the volume of urine and excretion of sodium returned to variable low levels never exceeding 40 mmol/24h.

Results of three infusion treatments given to eight subjects with nephrotic syndrome. Values are means (SD)

View this table:


Our study confirms previous reports that albumin causes negligible natriuresis in patients with strong sodium retention.1 Frusemide in sufficient dosage proved strongly natriuretic but we could not establish any potentiation of the frusemide effect by infusion of albumin. Only two studies have claimed a synergistic effect of combined albumin and frusemide, but both were retrospective and did not compare the effects of frusemide alone with those of combined frusemide and albumin in the same patients.2 3 Our study is the first to do so under controlled conditions. Albumin inside the renal tubules has been shown to bind frusemide and thus blunt its diuretic action.4 The increase in tubular albumin that occurs after infusion of albumin may have exerted some inhibitory action in our subjects, although because we applied a maximally effective dose such an effect is less likely. Our results do not support the use of albumin in the treatment of patients with the nephrotic syndrome.


  1. 1.
  2. 2.
  3. 3.
  4. 4.
View Abstract