Sodium glucose cotransporter-2 (SGLT2) inhibitors, originally developed for the treatment of type-2 diabetes, have shown efficacy in improving cardiovascular outcomes in patients with heart failure (both in heart failure with reduced and preserved ejection fraction), with and without diabetes. Additional evidence has been accumulating on the favorable effects of SGLT2 inhibitors in other cardiovascular pathologies.

Zelniker et al. reported that dapagliflozin reduced the risk of atrial fibrillation and atrial flutter by 19% among the 17,160 patients with type-2 diabetes who participated in the prospective multicenter randomized DECLARE-TIMI 58 trial (Dapagliflozin Effect on Cardiovascular Events-Thrombolysis in Myocardial Infarction 58) [1]. In the current issue of the Journal, Luo et al. report on a retrospective study assessing the effects of dapagliflozin, an SGLT2 inhibitor, on the outcomes of radiofrequency catheter ablation in patients with type-2 diabetes and atrial fibrillation [2]. Comparing a group that received dapagliflozin (n = 79) with a group that did not receive dapagliflozin (n = 247), they report that after a mean follow-up of 15.5 ± 8.9 months, the recurrence rate of atrial arrhythmia was reduced almost by half, with 27.8% of the patients in the dapagliflozin group versus 44.9% in the control group (p = 0.007). The association between dapagliflozin and lower risk for recurrence of atrial arrhythmia remained independent after multivariable Cox regression models. The results support the findings of two published studies. Kishima et al. conducted a prospective randomized controlled study comparing the effects of tofogliflozin (an SGLT2 inhibitor) and anagliptin (a dipeptidyl peptidase-4 inhibitor) in patients with type-2 diabetes after catheter ablation for atrial fibrillation [3]. Overall, 70 patients were analyzed including 30 with paroxysmal atrial fibrillation patients. After 12 months, recurrence of atrial fibrillation was detected in 24% of the tofogliflozin group versus 47% of the anagliptin group (p = 0.0417). Similarly, Abu-Qaoud et al. conducted a retrospective study, which included a total of 4450 patients with type-2 diabetes who underwent catheter ablation for atrial fibrillation [4]. They used a propensity-score matching and found that SGLT-2 inhibitor use was associated with a lower risk for cardioversion, use of new anti-arrhythmic agents, and need for repeat ablation.

The mechanism for the association between SGLT-2 inhibition and the lower risk of atrial fibrillation or reduced recurrence of atrial fibrillation after catheter ablation is unclear. SGLT-2 inhibitors alter the activation of numerous genes in the heart, attenuating cardiac hypertrophy, inflammation, fibrosis, and apoptosis; activating antioxidant enzymes; decreasing the expression of hypoxia markers; and altering cardiac cell metabolism [5, 6]; all can affect the (re-)development of AF. As the expression of SGLT2 in the heart is negligible, it is probably not directly related to the inhibition of SGLT2 in cardiac cells. While alteration of expression and/or activity of mediators secondary to SGLT2 inhibition in the kidney or other organs or the alteration of metabolic utilization cannot be ruled out, the fact that favorable effects of SGLT2 inhibitors are seen in cardiac cells in in vitro models suggests the potential existence of yet an unidentified alternative target. A recent study showed that the SGLT2 inhibitor empagliflozin can directly alter the metabolic profiles in the AC16 human cardiomyocyte cell lines, suggesting a direct SGLT-2 independent effect [7]. It has been postulated that inhibition of Na+/H+ exchanger (NHE-1) by SGLT2 inhibitors could be responsible for improved cardiac remodeling and reduced fibrosis with these agents. An in vitro study has shown that dapagliflozin attenuated lipopolysaccharide-induced increase in NHE-1 mRNA levels in mouse cardiac fibroblasts [8]. Activation of NHE-1 in cardiac fibroblasts is associated with extracellular matrix (ECM) remodeling, promoting fibrosis—a key pro-atrial fibrillation substrate [9].

The NHE-1 is involved in electrophysiological remodeling of atrial myocytes and AF [10, 11]. Enhanced NHE activity is linked to the increase of late INa and Na+ influx, which can prolong action potential duration (APD) and increase the likelihood of early afterdepolarizations (EADs). Additionally, activation of NHE1 can also lead to increased cytosolic Ca2+ concentration, subsequently activating Na+/Ca2+ exchanger (NCX) on sarcolemma or mitochondria respectively [12]. Activation of the sarcolemmal NCX is known to promote triggered activity by generating delayed afterdepolarizations (DADs). On the other hand, activation of the mitochondrial NCX can elevate the production of reactive oxygen species (ROS) by mitochondria. Elevated ROS levels could promote additional electrical and contractile remodeling by modifying the function of ion channels and Ca2+ handling proteins and activating NOD-like receptor 3 (NLRP3) inflammasome, which contributes to the pathogenesis of atrial fibrillation [9, 13]. Thus, inhibition of NHE-1 by SGLT2 inhibitor might restore the Ca2+ and Na+ homeostasis within atrial cardiomyocytes, attenuate the triggered activity induced by Ca2+ or Na+ overload, and normalize the inflammasome activity, thereby attenuating the atrial arrhythmogenicity. Another potential SGLT2-independent mechanism is the activation of adenosine monophosphate kinase (AMPK) by increasing the phosphorylation of AMPK (p-APMK). Increased p-AMPK levels by dapagliflozin have been described in vitro [8]. It is known that p-AMPK improves the metabolic function of the mitochondria and attenuates mitochondrial dysfunction during atrial fibrillation [14]. Long-term inhibition of NHE-1 by SGLT2 inhibitors might attenuate atrial remodeling and alleviate the substrate for atrial fibrillation through the inhibition of the abnormal downstream NHE-1 signaling (Fig. 1).

Schematic representation of the postulated mechanisms underlying the suppression of atrial fibrillation-promoting mechanisms mediated by an SGLT2 inhibitor. APD, action potential duration; AMPK, adenosine monophosphate kinase; DAD, delayed afterdepolarization; NCX, Na+/Ca2+ exchanger; NHE-1, Na+/H.+ exchanger-1; ROS, reactive oxygen species (Created with BioRender.com)

Despite the encouraging clinical results of various SGLT2 inhibitors in reducing atrial fibrillation risk, both challenges and research opportunities remain. Most existing studies are retrospective, which demonstrates the need for randomized and prospective studies to assess the impact of SGLT2 inhibitors on reducing the incidence or recurrence of atrial fibrillation in patients with or without heart failure, or type-2 diabetes. The BEYOND (Clinical BEnefit of sodium-glucose cotransporter-2 (SGLT-2) inhibitors in rhYthm cONtrol of atrial fibrillation in patients with diabetes mellitus) study is a multicenter, prospective open blinded-endpoint design, 1:1 randomized, and controlled study assessing the effects of SGLT2 inhibitors on the outcomes of rhythm control with anti-arrhythmic drugs and/or catheter ablation for atrial fibrillation in patients with type-2 diabetes [15]. A total of 716 patients will be enrolled. The primary outcome is recurrence rate of atrial fibrillation within a year documented on a 24-h Holter ECG. Thus, the jury is still out on the therapeutic effect of SGLT2 inhibitors on atrial fibrillation. Moreover, it is important to clearly demonstrate the mechanisms by which SGLT2 inhibitors prevent the development of atrial fibrillation, as this could help clinicians optimize clinical trial designs. With the advancement of single-cell sequencing and multi-omics, as well as the optimized protocols for generating human iPSC-derived cardiomyocytes, it becomes feasible to identify the cell-type–specific expression of SGLT2 in atrial tissue of patients, or animal models of atrial fibrillation. It is possible that, while SGLT2 expression is negligible in cardiac cells at baseline, de novo SGLT2 expression may exist in pathological conditions. Therefore, SGLT2 inhibitors could directly prevent atrial fibrillation–promoting mechanisms. On the contrary, if SGLT2 inhibitors have no direct effect on cardiac cells, this would further confirm that their role in ameliorating atrial arrhythmogenesis results indirectly from improved glucose metabolism and kidney function.

In conclusion, it is important to determine the mechanisms by which SGLT2 inhibitors could affect biological effects as upstream prevention treatment for AF and validation of these effects in randomized clinical trials.