Dapagliflozin effect on heart failure with prevalent or new-onset atrial fibrillation.

Abstract

This article refers to ‘Dapagliflozin and atrial fibrillation in heart failure with reduced ejection fraction: insights from DAPA-HF’ by J.H. Butt et al., published in this issue on pages XXX–XXX. Heart failure (HF) and atrial fibrillation (AF) are both heterogeneous conditions that affect millions of people worldwide.1 The overlap of clinical risk factors suggests that there may be similarities in remodelling processes ongoing in the heart.1 Advances in therapies have substantially improved the prognosis of HF patients with reduced ejection fraction (HFrEF) and AF separately, where treatment options in HF with preserved ejection fraction (HFpEF) are still limited.2, 3 Nonetheless, mortality persists high in both HFrEF and HFpEF, especially when combined with AF.1, 2 Therefore, new therapies targeting patients with the combination of HF and AF are eagerly needed. Sodium–glucose co-transporter type 2 (SGLT2) inhibitors, developed as a lowering-glucose therapy, were recently proven to prevent and treat HF.4 In the current issue of the Journal, Butt and colleagues add in timely manner evidence of the beneficial effects of dapagliflozin, an SGLT2 inhibitor, on outcomes of both with the combination of HFrEF and AF.5 The authors performed a sub-analysis of the DAPA-HF trial, and demonstrated that dapagliflozin, on top of guideline-recommended therapies, was as effective in patients with AF as compared to those without AF, to improve prognosis. In line with previous HF trials, approximately 40% of HFrEF patients had AF history or AF on electrocardiogram (ECG) at trial start, and the patients with the combination of HFrEF and AF were at increased risk of worsening HF and cardiovascular death.1, 6, 7 Most interestingly, dapagliflozin reduced the risk of HF hospitalization irrespective of AF presence. This is of great importance for all those patients with the combination of HFrEF and AF. Indeed, some of the established HF therapies demonstrate differential effects in HFrEF patients with and without AF. Angiotensin-converting enzyme inhibitors, angiotensin receptor blockers, neprilysin inhibitors, mineralocorticoid receptor antagonists, and implantable cardioverter defibrillators, all have beneficial effects on HFrEF irrespective of AF presence.3, 5 However, AF precludes effective delivery of biventricular pacing, and impacts the success of cardiac resynchronization therapy in HFrEF patients.8 Also, presence of AF negatively influences the prognostic benefit of beta-blockers in patients with HFrEF. Two meta-analyses of the major beta-blocker randomized controlled trials have shown that prognostic benefit of beta-blockers in HF patients with AF is less compelling than in HF patients with sinus rhythm.9, 10 However, beta-blockers do prevent new-onset AF or recurrent AF in HF.11 Reasons for absence of prognostic benefit are still uncertain, and seem unrelated to dosage and achieved heart rate.12 To this background, present finding of effectiveness of SGLT2 inhibitors in patients with HFrEF and AF is great news for this large patient population. Another interesting, but somewhat surprising, finding of Butt and colleagues is the absence of a beneficial effect of dapagliflozin on new-onset AF in patients with HFrEF.5 New-onset AF has a negative impact on HF hospitalizations, stroke, mortality, quality of life and exercise tolerance in patients with HF.2, 3 Therefore, prevention of new-onset or recurrent AF is an important treatment goal for HF patients. Heretofore, current guideline-recommended HF therapies that have beneficial prognostic effects also reduce new-onset AF (Figure 1).2, 3 Whether beta-blockers, angiotensin-converting enzyme inhibitors, angiotensin receptor blockers, neprilysin inhibitors, mineralocorticoid receptor antagonists, and cardiac resynchronisation therapy prevent new-onset AF via (i) direct atrial substrate effects, (ii) indirectly by optimizing the treatment of associated HF, or (iii) a combination of both direct and indirect effects, is not completely understood.3, 13 Explanations for absence of a relation between dapagliflozin and new-onset AF may relate to the mechanisms and study design. Indeed, a preventive effect on new-onset AF of SGLT2 inhibitors was expected based on their metabolic mechanisms. SGLT2 inhibitors result in the increase of insulin sensitivity, improvement of glycaemic control, lowering of systolic blood pressure, and weight loss, which are risk factors of both HF and AF.14 In addition, SGLT2 inhibitors may have direct atrial effects by improving mitochondrial respiration by reducing reactive oxygen species and the consumption of adenosine triphosphate.14, 15 Mitochondrial dysfunction has been suggested as a driver of atrial remodelling through the generation of reactive oxygen species which disturb cellular electrical activity, promote cardiomyocyte hypertrophy and interstitial fibrosis.14 Moreover, SGLT2 inhibitors are suggested to affect epicardial fat, which in turn causes local atrial inflammation, fatty infiltration and, potentially, arrhythmogenic substrate.14 Finally, SGLT2 inhibitors are suggested to increase serum magnesium with anti-arrhythmogenic effects.14 Some of the study design aspects to consider are the following. The DAPA-HF trial was set up as a HF trial, and diagnosis of AF was based on history or a single 10 s ECG at each study visit. The low intensity of AF monitoring may have led to missed AF diagnoses, as we have learned from studies using extended or continuous rhythm monitoring.6 This potential misdiagnosis of AF may not only have led to underdetection of all AF cases in those without AF at study start, but also may have led to an overestimation of new-onset AF which was in reality recurrence of AF. Needless to say, the sub-analysis was relatively underpowered due to the low number of patients with new-onset AF. The relative short follow-up of median 18 months for new-onset AF to occur may not have provided enough time for the protective metabolic mechanisms of SGLT2 inhibitors to settle in.6, 7 In addition, since dapagliflozin was administered on top of the guideline-recommended HF therapies, a potential effect of dapagliflozin may be overlooked by studying in DAPA-HF the added value of dapagliflozin. Butt and colleagues are to be congratulated for their contribution to the literature on the combination of HF and AF. They demonstrated elegantly that dapagliflozin reduces the worsening of HF and cardiovascular death in patients with HFrEF irrespective of AF status, but unexpectedly, could not demonstrate a reduction in new-onset AF. Given that fact that we should interpret sub-studies and post-hoc analyses with caution and use it only for exploratory and hypothesis-generating purposes, we need future studies investigating the effects of SGLT2 inhibitors on AF prevention and treatment, in the setting of (all types of) HF and without HF. Ultimately, further studies may help to optimize personalized treatment decisions in patients with the troublesome combination of HF and AF. Conflict of interest: none declared.