This is a randomized single-center clinical trial of the effect of spironolactone on top of standard treatment in patients with atrial fibrillation after sinus rhythm restoration on the recurrence of the arrhythmia, hospitalizations and on the changes in Gal-3 levels after 12 months.

After initial screening about the inclusion criteria, the patients were randomized in two groups. The active group received 25 mg Spironolactone on top of their usual therapy including antiarrhythmic medications, and the control group was treated according to the ‘usual (standart) care’ rhythm control.

The patients were followed up for 1 year and had 5 follow-up visits – at 14 days, 1 month, 3 months, 6 months, 9 months, and, finally, at 12 months.

The diagnosis of AF was accepted by ECG criteria during hospitalization or visit to the Emergency Department of the hospital. The type of AF was classified according to the ESC Guidelines on AF 2010 and 2016 (Camm et al. 2010; Camm et al. 2012; Kirchhof et al. 2016). Inclusion criteria were as follows: age more than 55 years, restored sinus rhythm after an episode of paroxysmal/persistent AF, signed inform consent. Exclusion criteria included the following: history of, clinical and echocardiographic evidence of chronic heart failure NYHA class III–IV; open heart surgery during the last 3 months for any indication; survivors of acute myocardial infarction and left ventricular dysfunction within 3 months of randomization; pregnancy; drug and alcohol abuse; presence of severe progressive concomitant disease with life expectancy less than 1 year; chronic kidney disease defined as serum creatinine more than 200 mcmol/l or eGFR less than 40 ml/min/1.73 m2; liver cirrhosis Child C; treatment with powerful CYP3A4 inhibitors or inductors; serum potassium levels >5 mmol/l at screening; hypersensitivity towards MRA; metabolic acidosis; known thyroid pathology with lab results consistent with hyper- or hypothyroidism.

At each visit, the patients were interviewed for episodes of recurrent arrhythmia, ECG proven by their physicians or at the follow-up visits or incidental visits to the EDs. Information about their vital status or other hospitalizations was also collected personally or by their relatives. The etiology was considered to be due to CVD or other reasons by the investigators (AK, YY, EG). The date of each episode was recorded, if known, or imputations of day 15^th for each month were done in case of unknown exact date of occurrence.

Blood for Gal-3 determination was collected at baseline and one year after. Ten mL of blood was drawn from the antecubital vein into BD Vacutainer SST II Advance Tubes. The blood was allowed to clot for 30 min at room temperature and then centrifuged at 1,500×g for 15 min at 4 °C. The separated serum was aliquoted into 1.5 mL polypropylene tubes, and stored at –80 °C until analysis. Samples with visible hemolysis were discarded from analysis.

Serum Gal-3 levels were determined using enzyme-linked immunosorbent assay kit for quantitative measurement (Galectin-3 AssayTM, REF# 12642-04, 12684 BG Medicine, Waltham, MA, USA) according to manufacturer’s instructions and were measured on StatFax 3200 microplate reader (Awareness Technology, Inc., USA). Calculation of results was based on 4 parameter logistic curve fit of the calibration curve and was performed with MikroWin 2000 ver. 4.31 software (Mikrotek Laborsysteme GmbH, Germany) and expressed in ng/mL units. The lower limit of detection (LoD) is 1.13 ng/mL, measurement range 1.4 to 94.8 ng/mL, average intra-assay CV: approximately 3.4% and average inter-assay CV: approximately 8.5%.

Standard 12-lead ECG was performed at each visit.

All continuous variables are presented as means ± standard deviation for relatively normally distributed and as median /interquartile range/ for these with deviation from normality. When approximately normal distribution is present, the independent variables are compared by Student’s t-test or ANOVA test in repeated measures in one patient. Because of skewed to the right distribution of Gal-3 values, we made a log transformation to improve the non-normal distribution. The paired t-test or one-sample t-test are applied for the differences in variables between the end and first visits. In case of lack of normality, nonparametric tests are also used like Mann-Whitney’s test. For categorical variables, absolute values and percentages are presented and the chi-square test or the exact Fisher’s test, when the expected cell numbers are smaller than 5 are used to test the null hypothesis. P-value <0.05 is used for significance testing.

For the occurrence of AF episodes during follow-up, the Kaplan-Meier curves were constructed with time to first event as dependent variable. Cox proportional hazard analyses were performed for the occurrence of AF events with different independent variables. First, univariate analysis was done and then, adjustment for major factors, like age (in categories <64 as reference, 64–67, 67–72, >72 years), sex (males vs females), diabetes (no vs DM vs impaired glucose tolerance), hypertension (yes/no), and duration of AF, was applied with backward selection modelling. The significance level 0.05 for keeping in the model and 0.1 for removing a variable from the model is used. Wald’s test for significance was done. The results are presented as hazard ratios (HR) with 95% confidence intervals (CI).

All analyses are performed on SPSS version 19 (SPSS, Texas, USA).

The project was approved by the local Committee of Medical Ethics of the University Hospital “St. Marina” Varna and complied with the Declaration of Helsinki. Informed consent was obtained in all patients.

In one patient spironolactone was stopped because of gynecomastia. In all patients creatinine and serum potassium were measured at visit 3.5 and 7. As expected, potassium levels were higher in spironolactone group. The mean difference between the groups is 0.2–0.36 mmol/l. All the measurements in patients, who has taken spironolactone ≥ 9 months are in the reference limit and the standard deviation does not exceed the upper normal limit, Fig. 2. Patients on spironolactone tend to have lower systolic blood pressure.

Figure 2. 

Levels of serum potassium; K – potassium.

The data about the influence of spironolactone on AF burden are inconsistent, especially in patients without heart failure. In TOPCAT (Treatment of Cardiac Function with an Aldosterone Antagonist) during a median follow-up of 3.3 years there was no differences in the rate of recurrences according to the treatment of spironolactone in patients with HFpEF (Neefs et al. 2020). Dabrowski et al showed that spironolactone in combination with beta-blocker has preventive effect on AF recurrences (Dabrowski et al. 2010). Other authors reported favourable effect of spironolactone on the fibrotic substrate in animal models, but there were no data about the number of recurrences (Lendeckel et al. 2010; Zhao et al. 2010). Tase et al. (2014) made a retrospective analysis of 1008 patients with similar characteristics as our population, divided in two groups – patients, who received spironolactone on top of amiodarone, propafenone or sotalol and patients on potassium supplementation on top of amiodarone, propafenone or sotalol (Tase et al. 2014). The AF recurrences 24 months before and so long after spironolactone initiation were detected. They found significant reduction of AF episodes in the spironolactone group. One possible reason for the lack of effect of spironolactone on AF recurrences is the great number of patients, who received ACE-inhibitors or angiotensin receptor blockers as antihypertensive agents, which contribute for RAAS blockade and diminish the difference between both groups. Another explanation is the relatively short duration of the study (one year), which period is not enough for spironolactone to influence the fibrotic process.

In TOPCAT spironolactone did not reduce the risk of the primary composite outcome of death from cardiovascular causes, aborted cardiac arrest or hospitalization for the management of HF in patients either with or without a history of AF at baseline (Cikes et al. 2018).

There are some preclinical data that spironolactone and modified citrus pectin inhibit the aldosterone-induced fibrosis (Calvier et al. 2013). In our study levels of Gal-3 increased in spironolactone group. This is in agreement with Aldo-DHF study in patients with HFpEF, in which Gal-3 increased in 6 months more rapidly in the spironolactone group (Edelmann et al. 2015). In contrast Deveci et al. (2015) found regression of Gal-3 after 6-months treatment with spironolactone on top of standard therapy from 39 ± 21 to 33 ± 22, р < 0.001, but in patients with reduced ejection fraction <35%. Patients with HFrEF usually have higher levels of aldosterone and may be this is why levels of Gal-3 decrease in such group an increases in non-HF patients, in which the levels of aldosterone are not elevated (Deveci et al. 2015).