SEARCH

SEARCH BY CITATION

Keywords:

  • aged;
  • congestive;
  • heart failure;
  • spironolactone

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. References

Aim

To assess how well heart failure patients tolerate spironolactone in routine clinical practice.

Design

Retrospective analysis of 226 patients attending a specialist heart failure clinic.

Results

One hundred and thirty of 226 (57.5%) patients tried spironolactone at least once. Forty-four of 130 (33.8%) discontinued spironolactone due to side-effects after a mean of 11.1 months; 59/141 (41.8%) trials of spironolactone resulted in at least one side-effect; therapy was stopped in 30/141 (21.3%) trials due to raised potassium or creatinine. Significant risk factors for raised potassium/creatinine were age and baseline potassium level.

Conclusions

Potentially serious side-effects are common despite appropriate use of spironolactone.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. References

The RALES (Randomized ALdactone Evaluation Study) [1] showed that spironolactone reduced the frequency of deaths and hospitalizations in patients with severe heart failure. It is unclear whether the findings can be generalized to the majority of patients suffering from severe heart failure, who are older and have more comorbidity [2].

Concern exists that older people, many of whom have impaired renal function and low aldosterone levels [3], may not tolerate spironolactone as well as the results of the RALES trial imply. Recent studies [4–7] suggest that life-threatening hyperkalaemia in patients taking spironolactone and angiotensin system blockers is common – more so in older people and those with Type 2 diabetes mellitus. It is unclear whether these findings were due to patient characteristics or to deviations from the RALES protocol [8].

The aim of this study was to ascertain how well spironolactone is tolerated in typical heart failure patients when the RALES prescribing guidelines are adhered to.

Methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. References

Data were collected from patients attending the Ninewells Heart Failure Clinic, which serves a mixed urban and rural population. Spironolactone was initiated only in New York Heart Association (NYHA) class III or IV patients with a history of NYHA class IV heart failure within the last 6 months. Potassium supplements and other potassium sparing diuretics were discontinued before starting spironolactone. Decisions to discontinue medication were taken on an individual basis by hospital physicians, clinic physicians and general practitioners (GPs), rather than strictly following the RALES criteria. Discontinuation of spironolactone for any reason is referred to as a ‘failed trial’ in this report.

Case records for all patients with chronic heart failure and objective evidence of left ventricular systolic dysfunction attending the clinic between 1 January 2001 and 31 December 2002 were examined. The study protocol was approved by the Tayside Local Research Ethics Committee and the local Caldicott Guardian. Data were anonymized and stored on a database prior to statistical analysis with SPSS version 11.5 (SPSS inc., Chicago, USA).

To compare tolerability between older and younger patients, the cohort was divided into patients aged < 75 and patients ≥ 75. Comparison of continuous variables was undertaken using Student's t-test. Dichotomous variables were analysed using χ2 test and Fisher's exact tests. Continuous variables were divided into quartiles for logistic regression analysis.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. References

Three hundred and ninety-four different patients attended during the above period. Three hundred and seventy-nine sets of notes (96.1%) were retrievable. Of 379 patients, 296 had a clinical diagnosis of chronic heart failure. Seventy of 296 patients had preserved LV function, leaving 226/296 patients with evidence of LV systolic dysfunction. Detailed data were abstracted from these 226 sets of notes and form the study population (Table 1).

Table 1.  Differences between older and younger patient subgroups
 All patients<75 years≥75 yearsP
  • *

    Number taking β-blockers when spironolactone first started, or if not started, at time of survey. Student's t-test for continuous variables; Pearson's χ2 for discrete variables.

Mean age, years73.365.881.1< 0.0001
Male sex156/226 (69.0%)85/116 (73.3%)71/110 (64.5%)0.156
Mild LV impairment47/225 (20.9%)21/116 (18.1%)26/109 (23.9%)0.555
Moderate LV impairment67/225 (29.8%)35/116 (30.2%)32/109 (29.4%)0.555
Poor LV function111/225 (49.3%)60/116 (51.7%)51/109 (46.8%)0.555
Frusemide dose equivalent (mg)60.355.464.90.141
Mean number of medications6.05.86.20.152
Mean Charlson comorbidity index2.942.643.25< 0.0007
Taking ACE inhibitor or AT2 blocker 209/226 (92.5%)109/116 (93.9%)100/110 (90.9%)0.384
Taking β-blockers*114/224 (50.9%)66/115 (57.4%)49/109 (45.0%)0.063
Stroke/transient ischaemic attack40/226 (17.7%)12/116 (10.3%)28/110 (25.5%)0.003
History of hypertension96/226 (42.5%)40/116 (34.5%)56/110 (50.9%)0.013
Diabetes50/226 (22.1%)31/116 (26.7%)19/110 (17.3%)0.087
Osteoarthritis, rheumatoid arthritis, joint replacement or osteoporotic fracture52/226 (23.0%)18/116 (15.5%)34/110 (30.9%)0.006

Tolerability of spironolactone

One hundred and thirty of 226 (57.5%) patients had tried spironolactone. Ninety-two of 130 (70.8%) were still taking spironolactone at the time of the survey, and 86/130 were taking spironolactone at the time of survey without a break (mean time 29.7 months, median 26 months). Thus 44/130 failed their first course of spironolactone (mean time to failure 11.1 months, median 8 months).

Of these 44 who failed a first trial of spironolactone, 11/44 (25.0%) went on to try spironolactone again. Five of 11 (45.5%) were still taking the drug at the time of survey (mean time on drug 12.8 months, median 14 months). Thus 6/11 (54.5%) failed their second trial of spironolactone (mean time to failure 1.2 months, median 1 month). Age did not influence the proportion tried on spironolactone (66/116 of those aged <75 years vs. 64/110 of those aged ≥75 years), but patients in the older age group were nonsignificantly more likely to stop taking spironolactone (20/66 of those aged <75 years vs. 24/64 of those aged ≥75 years).

Side-effects of spironolactone

Over 10% of patients discontinued spironolactone because of raised potassium, and another 10% stopped because of worsening renal function denoted by a rising creatinine (Table 2).

Table 2.  Side-effects of spironolactone
 All patients<75 years≥75 yearsRelative risk (95% CI)
  1. Includes all side-effects regardless of whether patient discontinued spironolactone as a result.

Hyperkalaemia 15/141 (10.6%)  8/74 (10.8%)  7/67 (10.4%) 0.97 (0.37, 2.52)
Raised creatinine 16/141 (11.0%) 4/74 (5.4%) 12/67 (17.9%) 3.31 (1.12, 9.78)
Hyperkalaemia or raised creatinine 30/141 (21.3%) 12/74 (16.2%) 18/67 (26.9%) 1.66 (0.86, 3.18)
Hypotension 7/141 (5.0%)  4/74 (5.4%)  3/67 (4.5%) 
Breast pain 5/141 (3.5%)  2/74 (2.7%)  3/67 (4.5%) 
Gynaecomastia  5/141 (3.5%)  4/74 (5.4%) 1/67 (1.5%) 
Hyponatraemia 3/141 (2.1%)  3/74 (4.1%)  0/67 (0%) 
N + V  3/141 (2.1%) 2/74 (2.7%)  1/67 (1.5%) 
Headache 1/141 (0.7%)  1/74 (1.4%)  0/67 (0%) 
Cramps 1/141 (0.7%) 0/74 (0%) 1/67 (1.5%) 
Impotence 1/141 (0.7%) 1/74 (1.4%)  0/67 (0%) 
Lightheaded  1/141 (0.7%)  0/74 (0%) 1/67 (1.5%) 
Tinnitus  1/141 (0.7%)  0/74 (0%)  1/67 (1.5%) 
Stomach pain  1/141 (0.7%)  1/74 (1.4%)  0/67 (0%) 
None 82/141 (58.2%) 43/74 (58.1%) 39/67 (58.2%)1.00 (0.76, 1.33)

Risk factors for raised creatinine and hyperkalaemia

Univariate analysis revealed advancing age (73.1 years vs. 78.2 years, P = 0.023) and higher pretreatment potassium level (4.48 vs. 4.16, P = 0.013) to be significant risk factors for discontinuation due to elevated potassium or creatinine. Diabetes mellitus [7/25 vs. 24/86, relative risk 1.00 (0.49–2.05)] and the dose of ACE inhibitor used (15.5 mg lisinopril vs. 12.9 mg lisinopril, P = 0.399) did not appear to be risk factors. Higher discontinuation rates in patients using nonsteroidal anti-inflammatory drugs [4/25 vs. 4/86, relative risk 3.44 (0.93–12.78)] and lower discontinuation rates in users of β-blockers [7/24 vs. 46/86, relative risk 0.55 (0.28–1.05)] did not reach significance; nor did baseline creatinine level (119 µmol/l vs. 133 µmol/l, P = 0.07). There was no significant difference in frusemide and spironolactone doses between those continuing with spironolactone (82.1 mg, 30.4 mg) and those discontinuing spironolactone (76.4 mg, 28.0 mg). In logistic regression analysis, only baseline potassium concentration retained predictive significance.

Of the 25 patients stopping their first course of spironolactone because of renal dysfunction, 13 stopped primarily for hyperkalaemia and 12 stopped primarily for rising creatinine. The mean baseline potassium in the hyperkalaemia subgroup was 4.54 mmol l−1, rising to a mean of 5.96 mmol l−1 at the time of stopping therapy. Eight of 13 (62%) had a potassium of ≥6.0 mmol l−1 at the time of stopping; 11/13 (85%) had a potassium of ≥5.5 mmol l−1. The mean baseline creatinine in the creatinine subgroup was 140.4 µmol l−1, rising to a mean of 197.2 µmol l−1 at the time of stopping therapy. Eight of 11 patients recorded a creatinine of >200 µmol l−1; 3/11 patients recorded a creatinine of <200 µmol l−1 and a rise from baseline of <50%. Four patients were hospitalized at the time of discontinuing spironolactone; the reasons for admission were chest pain, dehydration, worsening heart failure and diarrhoeal illness. Spironolactone was stopped by the GP in 8/25 (32%), by the heart failure clinic in 13/25 (52%), and by other hospital physicians in 4/25 (16%) of cases.

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. References

Our findings confirm that renal dysfunction and hyperkalaemia are common problems in heart failure patients taking spironolactone [4, 6, 9]. Of our patients, 20% discontinued spironolactone because of rising creatinine or high potassium, and in most cases this was due to major rises in creatinine or severe hyperkalaemia. Our clinic policy, together with the similarity of ACE inhibitor and spironolactone doses to the RALES study, suggests that differences in patient characteristics rather than non-adherence to prescribing indications is responsible for the high incidence of side-effects.

Severity of heart failure as measured by LV function or frusemide use did not predict discontinuation due to raised potassium or creatinine with spironolactone, in contrast to previous findings [6]. The better tolerability seen in patients taking β-blockers may reflect better tolerability for both drugs in patients with less severe heart failure. Patients with diabetes mellitus were no more likely to experience renal problems with spironolactone; previous studies have indicated an association [4, 5]. Although the overall frequency of side-effects was similar in older and younger patients, serious side-effects were commoner in older patients. Thirty percent of patients were unable to continue taking spironolactone long-term – higher than previously reported [8].

Our study has several weaknesses, being prone to biases inherent in retrospective casenote studies. We were unable to ascertain creatinine clearance in our patients. Small numbers have probably contributed to type 2 statistical errors in some of our analyses.

Although potentially serious side-effects are common, none of our patients died or required dialysis as a result of hyperkalaemia or renal dysfunction caused by spironolactone. Continual close monitoring is required to ensure that the benefits of spironolactone therapy are not offset by adverse effects, which may occur late after starting spironolactone. Future studies should examine how this can best be achieved.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. References
  • 1
    Pitt B, Zannad F, Remme WJ et al. The effect of spironolactone on morbidity and mortality in patients with severe heart failure. Randomized Aldactone Evaluation Study Investigators. N Engl J Med 1999; 341: 70917.
  • 2
    Lien CT, Gillespie ND, Struthers AD, McMurdo ME. Heart failure in frail elderly patients: diagnostic difficulties, co-morbidities, polypharmacy and treatment dilemmas. Eur J Heart Fail 2002; 4: 918.
  • 3
    King D. Diagnosis and management of heart failure in the elderly. Postgrad Med J 1996; 72: 57780.
  • 4
    Wrenger E, Muller R, Moesenthin M, Welte T, Frolich JC, Neumann KH. Interaction of spironolactone with ACE inhibitors or angiotensin receptor blockers: analysis of 44 cases. Br Med J 2003; 327: 1479.
  • 5
    Anton C, Cox AR, Watson RD, Ferner RE. The safety of spironolactone treatment in patients with heart failure. J Clin Pharm Ther 2003; 28: 2857.
  • 6
    Svensson M, Gustafsson F, Galatius S, Hildebrandt PR, Atar D. Hyperkalaemia and impaired renal function in patients taking spironolactone for congestive heart failure: retrospective study. Br Med J 2003; 327: 11412.
  • 7
    Berry C, McMurray JJ. Serious adverse events experienced by patients with chronic heart failure taking spironolactone. Heart 2001; 85: E8.
  • 8
    Bozkurt B, Agoston I, Knowlton AA. Complications of inappropriate use of spironolactone in heart failure: when an old medicine spirals out of new guidelines. J Am Coll Cardiol 2003; 41: 21114.
  • 9
    Schepkens H, Vanholder R, Billiouw JM, Lameire N. Life-threatening hyperkalemia during combined therapy with angiotensin-converting enzyme inhibitors and spironolactone: an analysis of 25 cases. Am J Med 2001; 110: 43841.