Hyperkalaemia and impaired renal function in patients taking spironolactone for congestive heart failure: retrospective study

BMJ 2003; 327 doi: (Published 13 November 2003) Cite this as: BMJ 2003;327:1141
  1. Morten Svensson, registrar1,
  2. Finn Gustafsson, senior registrar1,
  3. S⊘ren Galatius, senior registrar1,
  4. Per R Hildebrandt, head of department1,
  5. Dan Atar, professor (dan.atar{at}
  1. 1Frederiksberg University Hospital, Department of Cardiology and Endocrinology, Copenhagen, Denmark DK-2000
  2. 2Aker University Hospital, Division of Cardiology, Oslo, Norway
  1. Correspondence to: D Atar
  • Accepted 7 August 2003


Spironolactone reduces disease and death in patients with severe congestive heart failure and is well tolerated with regard to renal function and serum potassium concentrations.1 Guidelines recommend taking spironolactone in addition to angiotensin converting enzyme inhibitors and β blockers,2 3 but since spironolactone can lead to renal dysfuction or hyperkalaemia, we followed up a cohort of patients taking spironolactone to identify predictors of harmful effects.

Participants, methods, and results

We selected 125 consecutive patients from the congestive heart failure outpatient clinic of Frederiksberg University Hospital, Copenhagen (table).4 We included only patients with a left ventricular ejection fraction (LVEF) of no more than 45% or patients who were taking spironolactone. We started 65 patients on spironolactone; 60 patients were already taking spironolactone when they were referred. We measured blood electrolytes at baseline and then every two months. The study started in September 1999 and lasted 2 years. We analysed data using χ2 tests, Student's t tests, and multiple logistic regression.

Baseline characteristics of the heart failure population studied (n=125)

View this table:

At baseline, 93 (74%) patients were receiving potassium subplementation. We stopped subplements in 66 (71%) patients and gradually reduced dosages in the others. We observed each patient for a mean 370 days; total observation was for 73.0 patient years. We saw each patient a mean 11.1 times (mean 22.9 days between visits).

Mean peak serum creatinine concentration was 167.6 (SD 11.9) μmol/l, and mean peak potassium serum concentration was 5.0 (0.4) mmol/l. A total of 73 (58%) patients, had serum creatinine > 130 μmol/l, and 23 (18%) had > 220 μmol/l. Relative to baseline, 69 (55%) patients had their serum creatinine concentrations increase by 20%, 30 (24%) by 50%, and 11 (9%) by 100%. A total of 45 (36%) patients had potassium serum concentrations > 5 mmol/l, 21 (17%) patients > 5.5 mmol/l, and 13 (10%) > 6 mmol/l.

Patients taking spironolactone before referral did not differ significantly from those we started at the clinic in terms of severe hyperkalaemia (8.3% v 10.8%, P = 0.63) or azotaemia (23.3% v 24.6%, P = 0.81). We found a mean weight loss of 1.3 kg.

We logistically regressed azotaemia (defined as a 50% increase in serum creatinine concentration) with age, sex, New York Heart Association function class (NYHAFC), LVEF, and use of angiotensin converting enzyme inhibitors and β blockers. Age (odds ratio 1.74 (95% confidence interval 1.03-2.91) for each decade of age) and lower LVEF (0.94 (0.89 to 0.99) for each 10% increase) were independent risk factors for azotaemia.

For severe hyperkalaemia (serum potassium > 6.0 mmol/l) we added to the model baseline serum creatinine concentration and whether the patient was taking potassium subplementation. In this analysis the NYHAFC (3.36 (1.17 to 9.69) for each class) and lower LVEF (0.37 (0.15 to 0.95) for each 10% increase) were predictive factors.


Taking spironolactone for congestive heart failure is associated with considerably more frequent side effects than previously thought.1 Age, lower LVEF, and higher NYHAFC are predictors of hyperkalaemia and azotaemia.

Excessive diuresis might be an important cause of renal dysfunction while taking spironolactone. One way of dealing with renal dysfunction secondary to spironolactone may be to reduce doses of concomitant diuretics.

In the future, drugs such as eplerenone may replace spironolactone, but the risks of hyperkalaemia and renal insufficiency may be equally high.5 We found that adverse effects with spironolactone were common. We recommend that particular caution is taken in elderly people with LVEFs below 20%, potassium subplementation should be discontinued, changes in body weight should raise concern, doses in concomitant diuretic regimens may need adjustment, and continuous laboratory monitoring remains inevitable.


We thank P H Nielsen, H Bartholdy, I Henriksen, and L Flye Jensen.


  • Contributors FG, SG, PRH, and DA designed the study. MS collected the data. All authors did the analysis and formulated the results and discussion. MS wrote the first draft of the manuscript, and FG, SG, PRH and DA contributed to later editions. All authors reviewed the final version of the manuscript. DA is guarantor

  • Funding Merck, F Hoffmann-La Roche Ltd, Research Fund of the Copenhagen Hospital Corporation, Elisabeth M Schlinsog's Fund, Danish Hospital Foundation for Medical Research, and Danish Heart Foundation.

  • Competing interests None declared.

  • Ethical approval Ethics committee of the Frederiksberg and Copenhagen communities.


View Abstract