In the EMBODY trial, intragroup comparison revealed a significant improvement in parameters reflecting both sympathetic and parasympathetic nerve activities in the empagliflozin group, whereas intergroup comparison showed no significant difference between the empagliflozin and placebo groups. To be more precise, SDANN, r-MSSD, and HF, all of which reflect cardiac parasympathetic nerve activity, showed significant increases (improvement) in the empagliflozin group, but not in the placebo group. LF/HF ratio, which reflects cardiac sympathetic nerve activity, significantly decreased (improved) only in the empagliflozin group. HRT, which also reflects cardiac parasympathetic nerve activity, significantly improved in the empagliflozin group, but not in the placebo group. Whereas intergroup comparison showed no significant difference between the empagliflozin and placebo groups in the HRT.
As reflected in the results of SDNN and MIBG myocardial scintigraphy, both cardiac sympathetic and parasympathetic nerve activities improved in both the groups in the natural course of AMI. The EMBODY trial demonstrated that these effects might be enhanced by empagliflozin, which is an SGLT2 inhibitor. Conversely, we found no significant difference in LP as an indicator of depolarization or TWA as an indicator of repolarization abnormality. Although experimental data and clinical case reports have suggested that SGLT2 inhibitors improved cardiac sympathetic hyperactivity [16, 17], the present study, to the best of our knowledge, is the first trial to provide randomized clinical data demonstrating that empagliflozin improved both the cardiac sympathetic and parasympathetic activities in humans. The evidence connecting the autonomic nervous system to life-threatening arrhythmias and cardiovascular mortality is well established [18–20]. HRV is a physiological phenomenon characterized by beat-to-beat variations in cardiac cycle length, which is influenced by both sympathetic and parasympathetic autonomic tones. Abnormal HRV, both increased sympathetic and decreased parasympathetic activities, is presently considered a strong predictor for mortality and lethal ventricular arrhythmias in post-MI patients [21]. Several studies have reported that depressed HRV is associated with adverse outcomes in survivors of AMI [22]. Unlike depressed left ventricular ejection fraction, abnormal HRV predicts arrhythmic rather than nonarrhythmic mortality [23].
To date, HRT had been primarily examined in post-MI patients, and it is suggested that abnormal HRT is associated with increased mortality after MI [24]. Therefore, HRV and HRT are surrogate markers for fatal arrhythmias and sudden death, which were improved by empagliflozin in the present study.
Three major mechanisms were considered, which may underlie the improvement in HRV and HRT reflecting cardiac sympathetic and parasympathetic nerve activities with empagliflozin, an SGLT2 inhibitor.
The first mechanism is a hemodynamic effect. It is proposed that improved cardiac sympathetic nerve activity with SGLT2 inhibitors is due to reduced circulating intravascular volume through their osmotic diuresis and natriuresis [25]. This is reflected by an increase in hematocrit, which has also been found to be a key determinant of HF outcomes according to a recent exploratory analysis of the EMPA-REG OUTCOME trial [26]. It is hypothesized that the sustained reduction in intravascular volume and BP lead to a reduction in cardiac preload and afterload, respectively, thereby alleviating cardiac workload and improving LV function [25]. Such hemodynamic changes in intravascular volume and BP are observed without an increase in heart rate, suggesting that SGLT2 inhibitors reduce reflex sympathetic nerve hyperactivity or influence other neurohormonal pathways affecting the heart [27, 28]. In particular, SGLT2 inhibitors reduce BP after 24 hours; therefore, we believe that reduced BP led to the reduced cardiac sympathetic nerve activity [29, 30]. In the present study, the hematocrit levels were increased, BP was decreased, but heart rate did not change, which may result in the reduced cardiac sympathetic nerve activity.
The second mechanism is a myocardial energy supply effect. SGLT2 inhibitors have been reported to increase circulating levels of ketone bodies [31]. Ketones are freely taken up by myocardial cells, and may be a more efficient source of adenosine triphosphate for the failing heart compared with fatty acids [32]. Furthermore, it has been found that the utilization rate of ketone bodies is reduced during the course of myocardial ischemia [33]. Studies have suggested that increasing the use of ketone bodies, fatty acid, and branched-chain amino acids inhibited adenosine triphosphate (ATP) reduction, increasing ATP content in the myocardium in the empagliflozin group [34]. In the present study, the serum ketone bodies were also increased in the empagliflozin group, which may have resulted in the increased myocardium oxygen supply and decreased cardiac sympathetic nerve activity in patients with AMI being administered empagliflozin. In addition, SGLT2 inhibitors have been reported to increase in erythropoietin, which in itself may have cardioprotective effects, and to increase in hemoglobin, which may result in enhanced oxygen delivery to the myocardium [35, 36].
Furthermore, an emerging hypothesis suggests that SGLT2 inhibitors directly inhibit the myocardial sodium-hydrogen (Na+/H+) exchanger, thereby leading to increased mitochondrial calcium levels, improved mitochondrial function, reduced oxidative stress, and potentially reduced arrhythmias [37].
The third mechanism is a hepatic vagus nerve-mediated effect. The vagus nerve in the liver controls the activation of neurons in the rostral raphe pallidus (rRPa), which increases the cardiac sympathetic nerve activity, and heart rate [38]. The administration of SGLT2 inhibitors can reduce cardiac sympathetic nerve activity by reducing rRPa activity and in turn control the heart rate [39].
Furthermore, SGLT2 inhibitors reportedly reduced cardiac functioning and size of the infarction in a basic experimental model of AMI [40]. It has been reported that the activation of signal transducer and activator of transcription 3 (STAT3) is the underlying mechanism, which consequently exhibits antioxidative and anti-inflammatory activities [40]. This mechanism may be associated with the reduced cardiac sympathetic nerve activity. In the EMBODY trial, glycemic control levels were even in the two groups. We consider that the shift in blood glucose had very little impact on the improvement of sympathetic and parasympathetic activities with empagliflozin in the present study. Indeed, SGLT2 inhibitors have been shown to affect several other common modifiable risk factors and comorbidities associated with cardiovascular diseases, such as BW, renal function, uric acid level, and plasma lipid level [1–3]. Taken together, the results of the present study were comparable with those of previous studies [1–3].
In terms of safety, SGLT2 inhibitors are not devoid of undesirable effects related to marked glycosuria, such as genital infections, volume depletion, diabetic ketoacidosis (rare), and Fournier’s gangrene (extremely rare) [1–3]. In the present study, no such side effects were observed during the trial period, and no subjects discontinued the trial in the empagliflozin group.
Trial limitations
First, the trial period was 24 weeks. In the EMPAREG OUTCOME trial, reductions in CV death were observed during an early follow-up period (0–24 weeks), which may be due to decreased SCD with empagliflozin during this period. Thus, we hypothesized that the assessment period of 24 weeks adequately demonstrates the effects of empagliflozin on cardiac nerve activity as a surrogate of lethal ventricular tachyarrhythmias. Second, this trial was conducted only in the Japanese population. However, Japanese patients with coronary artery disease generally receive adequate conventional therapy, including statins. Therefore, it was possible to determine the exact therapeutic effects of empagliflozin against residual risk. β-blockers were not restricted during the trial period. Third, the baseline values of HRV and HRT are different between the two groups, and factors other than the allocation factor may be affected. However, the LF/HF ratio has improved with empagliflozin administration in the group with higher HRV, and there is no doubt that this reflects the improvement in cardiac sympathetic nerve activity. Finally, this study was conducted in a limited group called AMI, and it was considered that there was no significant difference in the intergroup comparison, because of spontaneous improvement in cardiac sympathetic nerve activity. Further studies are needed to evaluate whether SGLT2 inhibitors affect cardiac sympathetic nerve activity in other T2DM populations.