If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
1 This author takes responsibility for all aspects of the reliability and freedom from bias of the data presented and their discussed interpretation.
P. Sanders
Footnotes
1 This author takes responsibility for all aspects of the reliability and freedom from bias of the data presented and their discussed interpretation.
Affiliations
Centre for Heart Rhythm Disorders (CHRD), South Australian Health and Medical Research Institute (SAHMRI), University of Adelaide and Royal Adelaide Hospital, Adelaide, Australia
Despite advances in therapeutic interventions AF remains a progressive and symptomatic disease. Therefore, novel therapeutic interventions targeting the underlying arrhythmogenic substrate for AF is needed. Atrial fibrosis is an important component of the arrhythmogenic substrate of AF and may be initiated by aldosterone binding to the mineralocorticoid receptor. We hypothesized that aldosterone pathway blockade with mineralocorticoid receptor antagonists (MRA) reduces atrial fibrosis, and thus AF.
Methods
We searched OVID MEDLINE, OVID EMBASE and the Cochrane Central Register of Controlled Trials from inception to June 10th, 2016 for randomized controlled trials (RCT) and observational studies addressing MRA and providing information on AF occurrence. Two independent reviewers selected and appraised the data. We performed random-effects meta-analyses. Summary odds ratios (OR) with 95% confidence intervals (CI) were calculated.
Results
We included 14 studies, 5 RCT and 9 observational cohorts, with a cumulative number of 5332 patients (male: 74.9%, age: 65.3 years); 2397 (45.0%) received an MRA (spironolactone or eplerenone). During follow-up, 204 (8.5%) patients treated with MRAs, developed AF, compared to 547 (18.6%) patients, without MRA treatment. Meta-analyses showed a significant overall reduction of AF risk in MRA treated patients (OR: 0.48 CI: 0.38–0.60 p < 0.001), including a reduction of new-onset AF (OR: 0.52 CI: 0.37–0.74 p < 0.001) and recurrent AF (OR: 0.37 CI: 0.24–0.57 p < 0.001), but not post-operative AF (POAF) (OR: 0.60 CI: 0.33–1.09 p = 0.09).
Conclusions
MRAs significantly reduce new-onset AF and recurrent AF, but not POAF. MRA treatment can be considered an additive therapeutic strategy in AF.
Atrial fibrillation (AF) affects 1.5–2% of the European population, and is associated with severe comorbidities. It is associated with a five-fold increased risk of stroke, a three-fold increased incidence of congestive heart failure and a doubled mortality [
2012 focused update of the ESC Guidelines for the management of atrial fibrillation: an update of the 2010 ESC Guidelines for the management of atrial fibrillation.
in: Developed with the special contribution of the European Heart Rhythm Association. EHJ. 33. 2012: 2719-2747
]. Despite state-of-the-art therapeutic interventions AF remains a progressive, symptomatic disease. Therefore, novel therapeutic interventions targeting the arrhythmogenic substrate and preventing AF episodes are urgently needed.
The arrhythmogenic substrate of AF is driven by atrial fibrosis [
Cardiac fibrosis formation can be initiated by aldosterone binding to the mineralocorticoid receptor (MR). Aldosterone binding and subsequent cardiac fibrosis formation is indeed associated with increased AF propensity [
Impact of aldosterone antagonists on the substrate for atrial fibrillation: aldosterone promotes oxidative stress and atrial structural/electrical remodeling.
]. Moreover, among AF patients undergoing electrical cardioversion, a decrease in aldosterone plasma concentration was associated with longer SR maintenance [
In this systematic review and meta-analysis, we investigated the effect of MRAs on new-onset AF, post-operative AF (POAF) or recurrence of AF in a broad patient population with and without heart failure.
2. Methods
This systematic review was executed in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines [
]. The protocol was registered in the International Prospective Register of Systematic Reviews, registration number: CRD42015026461.
2.1 Search strategy
A medical information specialist (J.L.) systematically searched OVID MEDLINE, OVID EMBASE, The Cochrane Central Register of Controlled Trials and the non-MEDLINE subset of PubMed, from inception to June 10th, 2016. We used both controlled terms (i.e. MeSH-terms in MEDLINE) and free text terms for AF and MRA. Methodological filters were used to identify secondary and primary human studies. No language or date restrictions were applied. We cross-checked the reference lists and the citing articles of the identified relevant studies and adapted the search in case of additional relevant studies (Supplementary data).
2.2 Inclusion and exclusion criteria
We included studies among adults (≥18 years old), addressing the efficacy of MRAs (spironolactone and/or eplerenone), as intervention compared to a control arm, and using AF as outcome. The primary efficacy outcome was the risk of AF during the study-defined follow-up period. (Supplementary data).
The search identified 1004 unique abstracts. After review of titles and abstracts, 919 were excluded. A total of 29 titles were included based on a review of the abstract. Additionally, two studies of which only the abstract was available were included. After reviewing the full text, we excluded 17 studies (Fig. 1). AF detection was performed by a range of rhythm monitoring modalities, namely: telemonitoring, EKG, Holter and remote monitoring.
Fig. 1Flowchart of study selection. CENTRAL: The Cochrane Central Register of Controlled Trails.
We appraised the RCTs overall as low risk of bias. The critical appraisal of the observational cohort studies revealed a high risk of bias, which is due to retrospective collection of the data (Supplementary data).
3.2 Study population
The number of patients ranged between 74 and 1794 patients per study cohort, with 5332 patients in total [
Effect of combined spironolactone-beta-blocker ± enalapril treatment on occurrence of symptomatic atrial fibrillation episodes in patients with a history of paroxysmal atrial fibrillation (SPIR-AF study).
Association between renin-angiotensin-aldosterone system blockers and postoperative atrial fibrillation in patients with mild and moderate left ventricular dysfunction.
Angiotensin-converting enzyme inhibition or mineralocorticoid receptor blockade do not affect prevalence of atrial fibrillation in patients undergoing cardiac surgery.
Effectiveness of aldosterone antagonists for preventing atrial fibrillation after cardiac surgery in patients with systolic heart failure: a retrospective study.
Eplerenone and atrial fibrillation in mild systolic heart failure: results from the EMPHASIS-HF (Eplerenone in Mild Patients Hospitalization And SurvIval Study in Heart Failure) study.
]. All study cohorts included both patients with paroxysmal or persistent AF, except for the study cohorts of Ito et al. and Kim et al., who only included patients with persistent AF (n = 161 respectively n = 74 patients) [
]. In the RCTs 1243 (43.2%) patients were randomized to MRAs, whereas in the observational studies 1154 (47.0%) patients received MRAs. This was distributed over 1259 (52.5%) patients treated with spironolactone, and 966 (40.3%) treated with eplerenone. Concomitant treatment was optimal treatment for AF, heart failure and hypertension according to the applicable guidelines or as tolerated by the patient (Table 1). The total study cohort consisted of 74.9% males, and the mean age was 65.3 years. In 4400 patients the outcome was new-onset AF or POAF, whereas in 816 patients the outcome was recurrence of AF. A medical history of heart failure was described in 2866 patients (53.8%); hypertension was present in 3301 patients (61.9%).
Table 1Study characteristics of the included studies. AF: atrial fibrillation, MRA: mineralocorticoid receptor antagonist, POAF: post-operative atrial fibrillation, SR: Sinus rhythm.
Author and Year of publication
Country of inclusion
Number of Participants (Percentage treated with MRA)
Main inclusion criteria
Treatment specification (dosage)
Main outcome
Follow-up duration
Randomized-controlled trials
Gao et al. 2007
China
116 (50.0%)
Cardiomyopathy and CHF defined by Framingham diagnosis, NYHA III-IV, left ventricular ejection fraction < 45%
Spironolactone 25 mg
AF incidence
6 months
Dabrowski et al. 2010
Poland
164 (50.0%)
Symptoms of paroxysmal AF
Spironolactone 25 mg
Presence of symptomatic AF episodes on electrocardiogram
12 months
Marchetti et al. 2013
Italy
90 (50.0%)
Heart failure with systolic ventricular dysfunction and a first documented AF or atrial flutter and electrical cardioversion
MRA
Persistent AF incidence
3–24 months
Pretorius et al. 2012
USA
294 (50.0%)
Elective cardiac surgery and sinus rhythm
Spironolactone 25 mg
POAF
Until hospital discharge
Swedberg et al. 2014
Sweden
1794 (50.8%)
Congestive heart failure: NYHA II, age: >54 yrs., ejection fraction < 30%, ACE-inhibitors and beta-blockers in optimal dosing
Eplerenone 25 mg/50 mg
New-onset AF
21 months (median)
Observational prospective studies
Boldt et al. 2008
Germany
148 (27.0%)
Ventricular dysfunction (Ejection fraction 45%), who underwent ECV for AF lasting >24 h
MRA
At least 24 h of SR after cardioversion
At least 24 h.
Kim et al. 2009
Korea
74 (6.7%)
Persistent AF and external electrical cardioversion
Spironolactone
AF recurrence
13.2 (SD 11 months)
Özaydin et al. 2010
Turkey
269 (13.8%)
Primary coronary artery bypass and/or valve surgery and ejection fraction of ≤0.50
Spironolactone and ACE-inhibitor or angiotensin II receptor antagonists
POAF
Until hospital discharge
Observational retrospective studies
Brinkely et al. 2010
USA
171 (41.5%)
ICD implantation for primary prevention
Spironolactone
New-onset AF
21 months (mean)
Chung et al. 2015
Taiwan
1540 (50.0%)
End-stage renal diseases on dialysis, without history of cardiovascular disease
Spironolactone
New-onset AF
4.17 (IQ 25–75%: 1.5–8.63) years
Ito et al. 2013
Japan
161(34.1%)
Long-standing persistent AF and undergoing radio frequent catheter ablation
Eplerenone
Freedom of AF during follow-up
24 months
Paziaud et al. 2003
France
96 (21.9%)
Patients hospitalized in context of heart rhythm problems
Spironolactone 25 mg
SR
Not described
Simopoulos et al. 2013
Greece
332 (39.8%)
Elective cardiac surgery, no history of persistent AF or antiarrhythmic treatment
During follow-up, 204 (8.5%) patients who were treated with an MRA had an episode of AF, either new-onset or recurrence, compared to 547 (18.6%) patients, who were not treated with an MRA—yielding a RRR of 54.1% and an ARR of 10.1%, resulting in a NNT of 10. Meta-analysis showed a significant reduction of AF risk in MRA treated patients (OR:0.48 CI 95%:0.38–0.60, p < 0.001). To correct for potential bias of observational studies, we analyzed the RCTs and observational studies separately. Meta-analysis of RCTs resulted in a significant reduction of AF occurrence (OR:0.59 CI 95%:0.42–0.84, p = 0.003). This also applied to the meta-analysis of observational studies (OR: 0.42 CI 95% 0.32–0.55, p < 0.001) (Fig. 2). Furthermore, to correct for potential bias of retrospective studies, we analyzed the prospective and retrospective studies separately, which resulted in a similar significant reduction in MRA treated patients both in prospective and retrospective studies, (OR: 0.45 CI 95%: 0.34–0.59, p < 0.001) respectively (OR: 0.47 CI 95%: 0.30–0.75:, p = 0.002).
Fig. 2Forest plot of the meta-analysis of AF incidence and MRA treatment, including number of patients with AF during follow-up in the MRA and no MRA arm, with the calculated relative risk reduction. CI: Confidence interval. RRR: relative risk reduction.
Subgroup analysis of studies evaluating new-onset AF (1752 patients treated with MRAs and 1753 controls) showed a significant reduction of AF risk in MRA treated patients (OR: 0.52 CI 95%: 0.37–0.74, p < 0.001), leading to a RRR of 47.0%. The ARR was 3.2%—yielding a NNT of 31. (Fig. 3) [
Eplerenone and atrial fibrillation in mild systolic heart failure: results from the EMPHASIS-HF (Eplerenone in Mild Patients Hospitalization And SurvIval Study in Heart Failure) study.
]. However, the results of this study did not change the conclusion and exclusion of the study from the analysis yielded a similar significant reduction of AF risk in MRA treated patients (OR: 0.57 CI 95%: 0.38–0.84, p = 0.005).
Fig. 3Forest plot of meta-analysis of MRA treatment per subgroup: new-onset AF, POAF, recurrence of AF or AF in patients with heart failure. CI: Confidence interval.
Among a total of 572 patients with AF at baseline, 210 (36.7%) patients were treated with MRAs. Meta-analysis showed a significant reduction of AF risk in MRA treated patients (OR: 0.37 CI 95%: 0.24–0.57, p < 0.001), leading to a RRR of 57.9%. The ARR was 28.1%—yielding a NNT of 4 (Fig. 3) [
Effect of combined spironolactone-beta-blocker ± enalapril treatment on occurrence of symptomatic atrial fibrillation episodes in patients with a history of paroxysmal atrial fibrillation (SPIR-AF study).
]. When this study was excluded from the analysis a similar significant reduction of AF risk in MRA treated patients was seen (OR: 0.32 CI 95%: 0.18–0.57, p < 0.001). Recurrence of AF after cardioversion was evaluated in three studies [
]. Meta-analysis showed a significant reduction of AF recurrence after cardioversion in MRA treated patients (OR: 0.30 CI 95%: 0.15–0.57, p < 0.001).
3.7 Heart failure and atrial fibrillation risk
Additionally, we evaluated 857 patients with a history of heart failure. Patients in the MRA treatment arm had a significantly lower risk of AF compared to controls (OR: 0.46 CI 95%: 0.34–0.62, p < 0.001). The RRR was 37.1%. The ARR was 16.7%—yielding an NNT of 6 (Fig. 3) [
Effectiveness of aldosterone antagonists for preventing atrial fibrillation after cardiac surgery in patients with systolic heart failure: a retrospective study.
MRA treatment was not associated with a reduction of POAF in a selected study population of 895 patients undergoing cardiac surgery (OR: 0.60 CI 95%: 0.33–1.09, p = 0.09) (Fig. 3) [
Association between renin-angiotensin-aldosterone system blockers and postoperative atrial fibrillation in patients with mild and moderate left ventricular dysfunction.
Angiotensin-converting enzyme inhibition or mineralocorticoid receptor blockade do not affect prevalence of atrial fibrillation in patients undergoing cardiac surgery.
Effectiveness of aldosterone antagonists for preventing atrial fibrillation after cardiac surgery in patients with systolic heart failure: a retrospective study.
The Funnel plot and Egger's test (p-value: 0.001) suggested the presence of publication bias caused by publication of small study cohorts. This potential effect was analyzed post-hoc by excluding study cohorts of <100 patients [
]. This did not affect the main conclusion of our meta-analysis and resulted in a similar AF reduction in MRA treated patients (OR: 0.53 CI 95%: 0.42–0.66, p < 0.001). Additionally, after adjustment by Duval and Tweedie's trim and fill method, the odds ratio was similar (OR: 0.55 CI 95%: 0.43–0.70) (Supplementary data).
Based on the high potential of bias of the cohorts, of which only an abstracts was published, we have performed a sensitivity analysis excluding the two abstracts [
]. This did not alter our previous findings (data not shown).
4. Discussion
We performed an extensive systematic review and meta-analysis on the efficacy of MRA in AF reduction. We observed that: 1) MRAs were associated with a significantly lower AF risk than no MRA treatment. 2) This holds for patients with new-onset or recurrent AF, but not POAF. 3) The NNT to prevent one AF recurrence is 4, and seems larger in patients without than with heart failure.
The observed reduction of AF may partly result from antagonizing aldosterone binding to the MR resulting in less atrial fibrosis formation [
Impact of aldosterone antagonists on the substrate for atrial fibrillation: aldosterone promotes oxidative stress and atrial structural/electrical remodeling.
Impact of aldosterone antagonists on the substrate for atrial fibrillation: aldosterone promotes oxidative stress and atrial structural/electrical remodeling.
Impact of aldosterone antagonists on the substrate for atrial fibrillation: aldosterone promotes oxidative stress and atrial structural/electrical remodeling.
]. The renin-angiotensin-aldosterone system (RAAS) modulates oxidative stress by generation of ROS, such as nicotinamide adenine dinucleotide phosphate, leading to atrial fibrosis [
Impact of aldosterone antagonists on the substrate for atrial fibrillation: aldosterone promotes oxidative stress and atrial structural/electrical remodeling.
Impact of aldosterone antagonists on the substrate for atrial fibrillation: aldosterone promotes oxidative stress and atrial structural/electrical remodeling.
Effect of combined spironolactone-beta-blocker ± enalapril treatment on occurrence of symptomatic atrial fibrillation episodes in patients with a history of paroxysmal atrial fibrillation (SPIR-AF study).
]. A subgroup analysis of the RALES trial showed decreased levels of serum markers of cardiac fibrosis (PICP, PINP and PIIINP) upon aldosterone blockade [
Limitation of excessive extracellular matrix turnover may contribute to survival benefit of spironolactone therapy in patients with congestive heart failure: insights from the randomized aldactone evaluation study (RALES). Rales Investigators.
]. Hence, antagonizing aldosterone is suggested to decreases cardiac fibrosis. Future studies should focus on the exact origin of the decreased peptides.
Liu et al. described a 31% lower risk of AF in a review of 5 studies on MRA treated patients [
]. In our analysis, comprising 14 studies and 5332 patients, we found an even lower risk of new-onset or recurrent AF in MRA treated patients, both with and without heart failure. A better treatment heart failure rather than a direct anti-fibrotic effect cannot be excluded as cause for AF reduction. Nonetheless, the effect size of MRA on AF risk was a least as large in patients without heart failure.
MRAs significantly reduced new-onset AF, this was seen in both patients with and without heart failure, and in a large patient cohort without heart disease, but with end-stage renal disease at baseline. So currently there is limited data on new-onset AF reduction in people without heart disease, so called lone AF. It is too early to conclude that MRAs may benefit an otherwise healthy person, but further investigation is warranted.
MRAs did not significantly reduce POAF, which was mainly driven by the results of the RCT of Pretorius et al. [
Angiotensin-converting enzyme inhibition or mineralocorticoid receptor blockade do not affect prevalence of atrial fibrillation in patients undergoing cardiac surgery.
Angiotensin-converting enzyme inhibition or mineralocorticoid receptor blockade do not affect prevalence of atrial fibrillation in patients undergoing cardiac surgery.
]. Therefore, MRA treatment alone may not antagonize the acute, systemic inflammatory reaction in POAF. Prevention of POAF may entail more upstream anti-inflammatory therapy. Violi et al. demonstrated an absolute POAF risk reduction of 13% in patients treated with antioxidants [
]. In summary, MRAs may not halt the immediate inflammatory response in POAF, but may halt fibrosis formation, a secondary effect of inflammation and subsequently progression of AF [
Impact of aldosterone antagonists on the substrate for atrial fibrillation: aldosterone promotes oxidative stress and atrial structural/electrical remodeling.
Several aspects of the current systematic review and meta-analysis need to be taken into account when interpreting the results. The included studies vary in study design and clinically studies acquired different cohorts, which may lead to heterogeneity and multiple-testing errors. We have used random-effect models to adjust for potential statistical heterogeneity and performed subgroup analyses to correct for potential bias per study design. The effect of MRA was similar across different study designs and cohorts.
Additionally, a relatively large proportion of AF cases were derived from heart failure studies, which may have affected the results. However, we demonstrated convincingly that the AF reducing effect is as large, if not larger, in patients without heart failure, indicating an inherent effect on AF risk independent of heart failure presence. Because of limited related data, we could not perform subgroup analyses of other clinical parameters such as hypertension or concomitant medical treatment, which may influence the RAAS or AF.
Furthermore, the methods to detect AF during follow-up are not impermeable and heterogeneous across the studies. This is inherent to AF detection and may lead to underestimation of AF risk, however as this a limitation of all studies, this was considered a systemic error leading to correction of this potential form of bias. Therefore, we have not included methods and quality of AF detection in our study inclusion criteria.
The low incidence of AF, as in the study cohort of Chung et al. and Swedberg et al., could have underestimated the effects of MRA [
Eplerenone and atrial fibrillation in mild systolic heart failure: results from the EMPHASIS-HF (Eplerenone in Mild Patients Hospitalization And SurvIval Study in Heart Failure) study.
]. However, this would only strengthen the reported effects.
Finally, the small subgroups in the current analysis prevented us from performing subgroup analysis between spironolactone and eplerenone.
5. Conclusion
MRAs significantly lower the risk of both new-onset AF and recurrent AF in patients, but do not reduce POAF. The significant reduction is even more prominent in patients without a previous diagnosis of heart failure. MRAs could be regarded as a promising novel upstream therapy for prevention of AF in patients with underlying heart disease. Future research should focus on improving the identification of specific clinical parameters to select patients most likely to benefit from MRAs; a RCT will help to do so.
Conflict of Interest
The authors report no relationships that could be construed as a conflict of interest.
2012 focused update of the ESC Guidelines for the management of atrial fibrillation: an update of the 2010 ESC Guidelines for the management of atrial fibrillation.
in: Developed with the special contribution of the European Heart Rhythm Association. EHJ. 33. 2012: 2719-2747
Impact of aldosterone antagonists on the substrate for atrial fibrillation: aldosterone promotes oxidative stress and atrial structural/electrical remodeling.
Effect of combined spironolactone-beta-blocker ± enalapril treatment on occurrence of symptomatic atrial fibrillation episodes in patients with a history of paroxysmal atrial fibrillation (SPIR-AF study).
Association between renin-angiotensin-aldosterone system blockers and postoperative atrial fibrillation in patients with mild and moderate left ventricular dysfunction.
Angiotensin-converting enzyme inhibition or mineralocorticoid receptor blockade do not affect prevalence of atrial fibrillation in patients undergoing cardiac surgery.
Effectiveness of aldosterone antagonists for preventing atrial fibrillation after cardiac surgery in patients with systolic heart failure: a retrospective study.
Eplerenone and atrial fibrillation in mild systolic heart failure: results from the EMPHASIS-HF (Eplerenone in Mild Patients Hospitalization And SurvIval Study in Heart Failure) study.
Limitation of excessive extracellular matrix turnover may contribute to survival benefit of spironolactone therapy in patients with congestive heart failure: insights from the randomized aldactone evaluation study (RALES). Rales Investigators.
32577-3&id=fx1.jpg){kind=link}
32577-3&id=gr1.jpg){kind=link}
32577-3&id=gr2.jpg){kind=link}
32577-3&id=gr3.jpg){kind=link}