|Aldosterone blockade and left ventricular dysfunction: a systematic review of randomized clinical trials
|Ezekowitz JA, McAlister, FA
This review concluded that aldosterone blockers (spironolactone, eplerenone or canrenoate) given to selected people with heart failure or after acute myocardial infarction reduced the risk of death. However, patients should be assessed for risk factors before drug prescription due to some adverse effects. Some methods of the review were not described, but overall the authors’ conclusions appear reasonable.
To assess the effects of aldosterone blockers (spironolactone, eplerenone or canrenoate) on left ventricular dysfunction.
MEDLINE, Cochrane Central Register of Controlled Trials (CENTRAL) (to September 2007), EMBASE, International Pharmaceutical Abstracts, Science Citation Index Expanded (to September 2007) were searched, from inception to June 2008, unless otherwise stated. Search terms were reported. Seven relevant registers and four websites were checked for ongoing or unpublished trials. NLM Gateway, BioMed Central and OCLC PaperFirst were checked for meeting abstracts. References to included studies were screened and manufacturers were contacted. No language restrictions were applied.
Randomised controlled trials (RCTs) assessing the effects of spironolactone, eplerenone or canrenoate in people with symptomatic or asymptomatic left ventricular function were included. Eligible trials had to be of eight weeks or more duration and report on at least one of the clinical outcomes of interest. The comparator groups were placebo or active control. The primary outcome of interest was all-cause mortality. Secondary outcomes included all-cause hospitalisation, symptoms (New York Heart Association (NYHA) class, quality of life), functional capacity (exercise capacity, maximal oxygen uptake, six minute walk test), ejection fraction and safety outcomes (hyperkalaemia, renal failure, gynaecomastia).
The included trials were on people with acute myocardial infarction or those with heart failure. Where reported, inclusion criteria for included studies were: NYHA class range 2 to 4 in the majority of trials, but also included 1 to 3, or Killip 1 to 3; creatinine levels of less than 150 or less than 221μmol/L; serum potassium levels of less than 5.0 to 5.6μmol/L; and ejection fractions of less than 35% to less than 50%. The mean ages of included participants ranged from 54 to 70 years; between 29 and 92% were male. Where reported, 33 to 100% of disease was of ischaemic aetiology and the ejection fraction ranged from 19 to 68%. In the majority of included trials the comparator was placebo, but two used usual care, one usual care and metoprolol, and one additionally included a ramipril group (this group was not included in the analyses). The duration of trials ranged from two to 24 months.
One reviewer screened titles and abstracts for inclusion. Two reviewers independently assessed full text articles for inclusion; disagreements were resolved through consensus.
Assessment of study quality
Quality was assessed using the Jadad and Schultz methods.
The authors did not state how the validity assessment was performed.
Relative risk (RR) and 95% confidence intervals (CI) were calculated for dichotomous outcomes, and mean differences for continuous outcomes. The numbers and percentages of adverse events reported in each trial were tabulated. Where necessary authors were contacted for missing information.
The authors did not state how the data were extracted for the review, or how many reviewers performed the data extraction.
Methods of synthesis
Summary relative risk and weighted mean differences, together with their 95% confidence intervals, were estimated using random-effects models. The total number and percentages of adverse events were calculated.
Heterogeneity was assessed using the X2 test and the I2 statistics. Sensitivity analyses were undertaken based on comparison (placebo, active control), duration of treatment, year of publication or recruitment. Subgroup analyses were based on symptomatic versus asymptomatic left ventricular dysfunction, myocardial infarction versus others, and aetiology of heart failure, length of trial, year of publication or recruitment, and removal of unpublished trials.
Publication bias was assessed with Begg's and Egger's tests.
Results of the review
Nineteen RCTs (10,807 participants) were included; 14 assessed spironolactone, three canrenoate and three eplerenone. The majority of trials were small; nine had less than 50 participants and one trial, on people with acute myocardial infarction, contributed over 60% of the total participants. Jadad scores for trial quality ranged from 2 to 5 (out of 5 points); only five trials described clear allocation concealment.
Aldosterone blockade reduced all cause mortality (RR 0.80, 95% CI 0.74 to 0.87; I2=0%) and hospitalisation (RR 0.77, 95% CI 0.68 to 0.87; I2=38%; nine trials).
When results were stratified according to whether participants had acute myocardial infarction or heart failure, the effects on all cause mortality remained significant for both those with heart failure (RR 0.75, 95% CI 0.67 to 0.84; I2=0%; 14 trials) and those with acute myocardial infarction (RR 0.85, 95% CI 0.76 to 0.95; I2=0%; four trials); results for hospitalisation remained significant for those with heart failure (RR 0.73, 95% CI 0.63 to 0.84; I2=0%; seven trials), but not for those with acute myocardial infarction.
Major adverse events occurred more frequently in the aldosterone blockade group than the control group: hyperkalaemia 5.9% versus 3.0%, 17 trials; worsening renal function 8.9% versus 1.6%, 11 trials.
Other outcomes were reported inconsistently.
In sensitivity analyses, mortality was reduced in trials of 12 months or longer (RR 0.74, 95% CI 0.06 to 0.83; four trials), but there was no difference between trials lasting less than 12 months. The year of publication or recruitment and removal of unpublished studies made no significant difference to results.
There was no evidence of publication bias.
Aldosterone blockade was associated with a significant reduction in mortality in selected people with left ventricular dysfunction.
The aims of the review were clearly stated in terms of study design, intervention and population. The search covered a wide range of sources, including unpublished studies, and was not limited by language; this reduced the likelihood of publication bias affecting the review. This appeared to be confirmed by the test results for publication bias. The methods used for data extraction and quality assessment were not described, so it was not possible to evaluate whether these were aimed at reducing reviewer error or bias.
Trial quality was assessed using standard scoring systems, but the authors did not appear to use trial quality to inform the analyses. The methods of combining results appeared appropriate. Heterogeneity was investigated, although some of the subgroup analyses suggested in the methods section were not reported in the results, whereas different subgroup analyses were reported. As the authors emphasised, the participants in the included trials were those at lower risk of adverse effects of aldosterone blockers, affecting the generalisability of the results.
Whilst there are some concerns about the lack of information on some of the methods of the review, overall the authors' conclusions appear reasonable.
Implications of the review for practice and research
Practice: The authors stated that the safety profile of aldosterone blockade should be considered before prescribing, but used correctly it should be considered a key medication.
Research: The authors stated that further study in people with less severe symptoms or preserved systolic function is warranted.
Alberta Heritage Foundation for Medical Research; Canadian Institutes of Health Research Randomized Controlled Trials Program.
Ezekowitz JA, McAlister, FA. Aldosterone blockade and left ventricular dysfunction: a systematic review of randomized clinical trials. European Heart Journal 2009; 30(4): 469-477
Subject indexing assigned by NLM
Aged; Cause of Death; Female; Heart Failure /drug therapy /mortality /physiopathology; Hospitalization; Humans; Male; Middle Aged; Mineralocorticoid Receptor Antagonists /adverse effects /therapeutic use; Randomized Controlled Trials as Topic; Spironolactone /therapeutic use; Stroke Volume /drug effects; Treatment Outcome; Ventricular Dysfunction, Left /drug therapy /mortality /physiopathology
Date bibliographic record published
Date abstract record published
This is a critical abstract of a systematic review that meets the criteria for inclusion on DARE. Each critical abstract contains a brief summary of the review methods, results and conclusions followed by a detailed critical assessment on the reliability of the review and the conclusions drawn.