Pharmacological inhibition of SK‐channels with AP14145 prevents atrial arrhythmogenic changes in a porcine model for obstructive respiratory events

Abstract

Background

Obstructive sleep apnea (OSA) creates a complex substrate for atrial fibrillation (AF), which is refractory to many clinically available pharmacological interventions. We investigated atrial antiarrhythmogenic properties and ventricular electrophysiological safety of small‐conductance Ca²⁺‐activated K⁺ (SK)‐channel inhibition in a porcine model for obstructive respiratory events.

Methods

In spontaneously breathing pigs, obstructive respiratory events were simulated by intermittent negative upper airway pressure (INAP) applied via a pressure device connected to the intubation tube. INAP was applied for 75 s, every 10 min, three times before and three times during infusion of the SK‐channel inhibitor AP14145. Atrial effective refractory periods (AERP) were acquired before (pre‐INAP), during (INAP) and after (post‐) INAP. AF‐inducibility was determined by a S1S2 atrial pacing protocol. Ventricular arrhythmicity was evaluated by heart rate adjusted QT‐interval duration (QT‐paced) and electromechanical window (EMW) shortening.

Results

During vehicle infusion, INAP transiently shortened AERP (pre‐INAP: 135 ± 10 ms vs. post‐INAP 101 ± 11 ms; p = .008) and increased AF‐inducibility. QT‐paced prolonged during INAP (pre‐INAP 270 ± 7 ms vs. INAP 275 ± 7 ms; p = .04) and EMW shortened progressively throughout INAP and post‐INAP (pre‐INAP 80 ± 4 ms; INAP 59 ± 6 ms, post‐INAP 46 ± 10 ms). AP14145 prolonged baseline AERP, partially prevented INAP‐induced AERP‐shortening and reduced AF‐susceptibility. AP14145 did not alter QT‐paced at baseline (pre‐AP14145 270 ± 7 ms vs. AP14145 268 ± 6 ms, p = .83) or QT‐paced and EMW‐shortening during INAP.

Conclusion

In a pig model for obstructive respiratory events, the SK‐channel‐inhibitor AP14145 prevented INAP‐associated AERP‐shortening and AF‐susceptibility without impairing ventricular electrophysiology. Whether SK‐channels represent a target for OSA‐related AF in humans warrants further study.

Obstructive respiratory events, shortening of atrial refractoriness and efficacy of AAD: Obstructive respiratory events may be associated with venous preload, arousal, thoracic pressure swings and asphyxic blood gas changes (hypoxia and hypercapnia). These pathophysiological elements may contribute to shortened atrial refractoriness in the setting of OSA. While established AADs could not blunt apnea‐related shortening in atrial refractoriness, AP14145, as a novel SK‐channel inhibitor could. Nevertheless, SK‐channel involvement in OSA‐related AF remains putative and further investigations are warranted. AAD, atrial antiarrhythmic drugs, OSA, obstructive sleep apnea.

1. INTRODUCTION

Obstructive sleep apnea (OSA) is present in up to 78% of patients with atrial fibrillation (AF) ¹ , ² and is associated with reduced efficacy of rhythm control interventions, such as catheter‐based and pharmacological treatment approaches. ³ , ⁴ , ⁵ The effect of novel antiarrhythmic treatment targets on OSA‐related atrial arrhythmogenic mechanisms, which may critically contribute to the AF substrate in a magnitude of AF patients, is not routinely studied during drug development processes.

Although OSA is a chronic disease associated with a multitude of pathophysiological changes, which may set the foundation for a structural substrate of AF, single nights of OSA are suspected to be a crucial part of OSA‐related AF. In a recent observational study, OSA patients who had undergone a night of severe sleep apnea had an increased risk for AF episodes during the following day compared to nights with less pronounced OSA. ⁶ AF risk can be augmented by not only single nights of OSA, but also by single apneic episodes. Previously, we demonstrated in a porcine model for OSA that single obstructive respiratory events are associated with transient hypoxemia, intrathoracic pressure changes and an arousal‐related sympatho‐vagal activation. ⁷ , ⁸ These events resulted in acute and transient shortening of atrial effective refractory period and an increase in AF inducibility, creating an apnea associated dynamic AF substrate. ⁸ , ⁹ Such highly dynamic and short‐term arrhythmogenic changes were resistant to available antiarrhythmic drugs such as amiodarone, sotalol, dofetilide and I _Kur inhibition (AVE0118). ¹⁰ Moreover, autonomic nervous system interventions, such as renal denervation, low‐level baroreceptor stimulation or atenolol, could only partially prevent atrial refractory period shortening and AF inducibility. ¹¹ , ¹²

A novel pharmacological treatment for AF that has not yet been tested on OSA‐associated arrhythmogenic substrates involves targeting of the small‐conductance Ca²⁺‐activated K⁺ (SK)‐channels. SK‐channels are expressed in the atria across many mammalian species and have been demonstrated to contribute to atrial repolarization in humans. ¹³ Inhibition of SK‐channels is associated with prolongation of the atrial action potential and effective refractory period, resulting in a reduction of AF‐duration. ¹³ , ¹⁴ As SK‐channels predominantly play a functional role in the atria, these channels provide a unique opportunity for a rather atrial‐specific antiarrhythmic therapy. This is especially apparent in the case of OSA, since OSA itself has been linked to increased ventricular arrhythmia susceptibility. ⁷ , ¹⁵

To investigate the relevance of these SK channels in OSA, we utilized an established pig model for obstructive respiratory events, which is induced by applying intermittent negative upper airway pressure (INAP). Using this model, we tested whether pharmacological inhibition of SK‐channels by AP14145 intervention protects against INAP‐associated arrhythmia risk. To investigate drug safety, we scrutinized the impact of AP14145 on ventricular repolarization in this disease model, especially since OSA is also associated with ventricular arrhythmia risk.

2. METHODS

2.1. Animals

All animal experiments were performed in accordance to the Danish law for the protection of animals. The investigation conforms to the guidelines from Directive 2010/63/EU of the European Parliament on the protection of laboratory animals and aligns with the ARRIVE‐guidelines. The study has been reviewed and approved by the regional Animal Welfare Inspectorate (Ministry of Environment and Food of Denmark: license# 2018‐15‐0201‐01608). Twelve female Danish landrace pigs (50 kg) were housed, two per pen under standardized conditions (12 h dark/light cycle) with tap drinking water ad libitum and feeding with a regular diet twice a day.

2.2. Anesthesia and euthanasia

Pigs were premedicated with Zoletil (250 mg dry Tiletamine + Zolazepam, 6.5 ml Xylazine 20 mg/ml, 1.25 ketamine 100 mg/ml, 2.5 ml Butorphanol 10 mg/ml and 2 ml methadone 10 mg/ml) 1 ml/10 kg intramuscularly. Throughout the experiment, pigs were anesthetized with alpha‐chloralose 4% (0.4 ml/kg intravenous loading bolus, 0.1 ml/kg/h maintenance) and fentanyl (5 μg/kg/h intravenously), which left them in a stable sedation that still allowed for them to breathe spontaneously. Throughout the experiment, 1 ml Zoletil was administered intramuscularly every hour. All experiments were nonrecovery. For euthanasia, pigs received an intravenous bolus of pentobarbital (160 mg/kg).

2.3. Simulating obstructive respiratory events—application of INAP

INAP was applied by a custom‐made negative pressure device, with a 50 L pressure container and a vacuum pump controlled by a manometer (Manometer Type 831, WIKA, Alexander Wiegand SE & Co. KG) as described previously. ⁷ , ⁸ , ⁹ , ¹⁰

2.4. Apnea‐protocol

INAP was maintained for 75 s followed by a 10‐min resting period. Apart from the first INAP, all other INAP interventions were applied while determining atrial effective refractory period (AERP). The order of apneas and drug interventions are depicted in Figure 1A. Arterial blood gas samples were taken before and after each INAP and analyzed by “ABL90 FLEX blood gas analyzer” (Radiometer).

Figure 1.

Experimental protocol and representative traces: (A) Overview of experimental protocol. (B) Representative traces throughout 75 s of INAP. Start and end of INAP indicated by arrows in the traces for “Upper airway pressure”. Thiry seconds after INAP‐onset (start in airway pressure fluctuation and first arrow) an atrial effective refractory period was determined. Here, a measurement related coupling stimulus (S2) initiated atrial fibrillation (arrow in MAP channel). SpO₂ dropped with a slight delay after INAP‐onset, but recovered within 2 min after INAP. Atrial MAP, atrial monophasic action potential recording; CVP, central venous pressure; ECG—a‐p, electrocardiogram anterior–posterior axis; INAP, intermittent negative upper airway pressure; SpO₂, oxygen saturation.

Vehicle (saline 0.9 mg/ml and PEG400 [Polyethyleneglycol] ratio 1:1; sterile filtered) was administered as a loading dose of a volume of 1 ml/kg (over 1 min) and a succeeding vehicle infusion of 2 ml/kg/h (total of 1 ml/kg over 30 min). For AP14145 administration loading dose was 5 mg/kg AP14145 (volume of 1 ml/kg over 1 min) and a succeeding infusion of 5 mg/kg AP14145 (infusion speed of 2 ml/kg/h and a volume of 1 ml/kg over 30 min). Blood samples were analyzed to verify stable and constant AP14145 compound levels (Supporting Information: Figure S1).

2.5. Atrial and ventricular electrophysiology

AERP measurements were performed before (pre‐INAP), during (INAP) and shortly after INAP (post‐INAP). AERP measurements were performed with eight stimuli (S1), with a pulse width of 2 ms, and an amplitude of 3× rheobase and 500 BCL ms, followed by an extra‐stimulus (S2) applied with 10 ms increments starting from: 2/3 of the pre‐INAP‐AERP for INAP and −40 ms of INAP for post‐INAP.

Atrial responses to the pacing protocol were visualized with right atrial endocardial monophasic action potential (MAP) recordings. MAP signals were only used for analysis if a regular baseline and amplitude of the MAPs was maintained and if the shape of the MAP was stable during the INAP procedures. Right atrial MAP duration was evaluated from 90% repolarization (APD90) during regular pacing (BCL 500 ms). The occurrence of S2‐induced atrial tachycardia was used as a parameter for AF‐inducibility. After the first S2 was captured during the AERP measurement procedure, four more S2 stimuli were applied while increasing the coupling interval by 10 ms for the subsequent S1–S2 procedure. AF inducibility was only determined for pre‐INAP and post‐INAP. If AF persisted for 3 min, it was DC‐converted. QT‐interval duration (QT‐paced) and electromechanical window (EMW) calculation was assessed during fixed atrial pacing, and measured as described previously. ⁷

2.6. Statistics

Data are expressed as mean ± standard error of the mean (SEM). For assessment of statistical significance between all three time points (e.g., pre‐INAP vs. INAP vs. post‐INAP), a paired one‐way analysis of variance followed by Tukey's multiple comparison test was applied. For comparison of two time points, such as drug effect and blood gas analysis, a paired t‐test was used. p Values are considered as statistically significant if <.05. Statistical analysis was carried out using “GraphPad PRISM Version 8.0.1.”

3. RESULTS

3.1. Effects of INAP on blood gases and central venous return

As a simulation of obstructive respiratory events, INAP was applied to spontaneously breathing pigs. Throughout INAP, upper airway pressure dropped repeatedly (representative traces in Figure 1B), and was associated with decreased arterial pH (pre‐INAP 7.41 ± 0.01 vs. INAP 7.36 ± 0.01; p = .03), increased arterial CO₂ partial pressure (pre‐INAP 6.65 ± 0.26 kPa vs. 7.53 ± 0.39 kPa; p = .02), and decreased arterial O₂ partial pressure (pre‐INAP 12.57 ± 0.48 kPa vs. 6.27 ± 0.51 kPa; p < .001). All respiratory parameters were reversible upon recovery. Lactate levels were not influenced by INAP (pre‐INAP 1.37 ± 0.15 mmol/l vs. 1.34 ± 0.21 mmol/l; p = .53). During normal breathing, central venous pressure (CVP) fluctuated according to breathing (representative traces in Figure 1B), with mean values of 3 ± 1 mmHg. INAP was associated with an increase in CVP (pre‐INAP 3 ± 1 mmHg vs. post‐INAP 8 ± 2 mmHg; p < .001). INAP‐induced changes of CVP remained stable throughout the protocol and in the presence of AP14145 (after AP14145: pre‐INAP 2.6 ± 1 mmHg vs. post‐INAP 6.6 ± 2 mmHg; p = .004).

3.2. Atrial electrophysiological changes induced by INAP and effects of AP14145

INAP was associated with progressive shortening in AERP (pre‐INAP: 135 ± 10 ms; INAP: 119 ± 12 ms; post‐INAP 101 ± 11 ms), with the shortest values occurring after INAP (pre‐INAP vs. post‐INAP, Figure 2A). Accordingly, atrial APD90 shortened progressively and similarly demonstrated the lowest values in post‐INAP (pre‐INAP: 132 ± 4 ms; INAP: 124 ± 7 ms; post‐INAP 104 ± 9 ms; Figure 2B).

Figure 2.

Atrial arrhythmogenic effects: INAP associated progressive shortening of atrial effective refractory period (AERP) (A) and action potential duration at 90% repolarization (APD90) during vehicle (B). After administration of AP14145, neither AERP (C) nor APD90 (D) shortened INAP‐related. Intermittent negative upper airway pressure (INAP). Data are expressed as mean ± SEM. For statistical analysis, a one‐way analysis of variance followed by Tukey's multiple comparison test was applied.

AP14145 increased AERP by 41 ± 7 ms at pre‐INAP (vehicle vs. AP14145; p < .001) without statistically significant changes in atrial APD90 (vehicle vs. AP14145 + 16 ± 8 ms; p = .12), indicating post‐repolarization refractoriness (PRR: AERP—APD90: 23 ± 9 ms). In contrast to baseline measurements, INAP did not alter AERP (Figure 2C) or atrial APD90 (Figure 2D) after AP14145 administration. INAP shortened AERP in the presence of AP14145 by 3 ± 5 ms, whereas the refractory period only shortened by 34 ± 8 ms under baseline conditions (p = .008).

As a reference point, vehicle treated time controls (n = 3) responded with a comparable shortening of AERP to INAP application indicating no accumulating effects by the INAP protocol (pre‐INAP 131 + 4 ms, INAP 96 + 12 ms, post‐INAP 78 + 13 ms).

3.3. INAP and AF‐inducibility

INAP application was associated with increased AF‐inducibility (representative traces in Figure 3A). Whereas pre‐INAP/ + Vehicle AF‐inducibility was 11% ± 8%, AF‐inducibility increased throughout INAP to 48 ± 11% (Figure 3B; p = .03). INAP associated AF‐inducibility in the presence of AP14145 was 5 ± 5%, which represented an overall reduction in AF‐inducibility by 43 ± 12% (Figure 3B,C; p = .012).

Figure 3.

Atrial fibrillation inducibility by extra‐stimuli. (A) Representative traces of an atrial refractory measurement with a train of eight S1 stimuli and a shorter coupling stimulus (S2), which induced atrial fibrillation (AF). (B) AF‐inducibility pre‐INAP and INAP during vehicle and AP14145. (C) Comparison of INAP‐related AF‐inducibility before and after AP14145. Intermittent negative upper airway pressure (INAP). Data are presented as mean ± SEM. For statistical analysis, a paired t‐test was applied.

3.4. Effects of AP14145 on INAP‐induced ventricular electrophysiological changes

INAP was associated with dynamic INAP‐induced QT‐interval durations (QT‐paced). INAP transiently prolonged QT‐paced intervals (pre‐INAP 270 ± 7 ms vs. INAP 275 ± 7 ms; p = .038) and shortened progressively the EMW (pre‐INAP 80 ± 4 ms; INAP 59 ± 6 ms, post‐INAP 46 ± 10 ms; Figure 4A,C). AP14145 did not change QT‐paced (pre‐AP14145 270 ± 7 ms vs. AP14145 268 ± 6 ms, p = .83) and dynamic changes in QT‐paced due to INAP remained (Figure 4B). However, EMW‐shortening during post‐INAP was blunted, depicting overall longer EMW‐values than in the vehicle post‐INAP (Vehicle post‐INAP 46 ± 10 ms vs. AP14145 post‐INAP 79 ± 8 ms; p = .027).

Figure 4.

Ventricular electrophysiological effects: Dynamic QT‐interval duration while fixed atrial pacing (QT‐paced) pre‐INAP, INAP and post‐INAP, during vehicle (A) and AP14145 (B). Changes in ventricular electromechanical window (EMW) pre‐INAP, INAP and post‐INAP, during vehicle (C) and AP14145 (D). Intermittent negative upper airway pressure (INAP). Data are expressed as mean ± SEM. For statistical analysis, a one‐way analysis of variance followed by Tukey's multiple comparison test was applied.

4. DISCUSSION

Acute obstructive respiratory events contribute to a complex substrate for AF, which remains difficult to control, and therefore requires novel treatment strategies. In this study, we demonstrated that pharmacological inhibition of SK‐channels by AP14145 protects against INAP‐induced atrial proarrhythmogenic changes. These changes are not associated with prolonged ventricular repolarization or with a potentiated disturbance of electromechanical interaction, which is represented by a shortened EMW.

4.1. Implications for atrial antiarrhythmic drug development in OSA

A recent meta‐analysis aiming to quantify global OSA prevalence estimated that approximately one billion people worldwide are affected by OSA. ¹⁶ Even though endeavors are increasing to diagnose OSA quickly and before secondary cardiovascular diseases to OSA can manifest, it remains a disease that is still broadly underdiagnosed, which poses a significant burden on public health. ¹⁷ , ¹⁸

OSA has been reported to be present in up to 78% of AF patients. ² , ¹⁹ Aside from structurally induced changes due to chronic OSA, ²⁰ such as atrial enlargement and increased fibrosis formation, there is a growing body of evidence supporting relevant acute and transient arrhythmogenic changes induced by single apneas or single nights of severe OSA. ⁶ , ⁸ In a recent observational study in OSA‐patients, the risk of having AF for at least 5 min during the day was increased and linked to suffering from severe OSA during the previous night, implying the importance of short‐term OSA. ⁶ The transient and predominant nocturnal arrhythmogenic changes have been studied in detail in pigs, in which single obstructive respiratory events were associated with a transient decrease in atrial refractoriness. ⁸ This has been proven to be resistant towards a broad spectrum of established antiarrhythmic compounds. ¹⁰ When considering arrhythmogenic effects of OSA on ventricular repolarization, atrial antiarrhythmic treatment becomes even more challenging, since a majority of atrial antiarrhythmic drugs (AAD) are associated with an impact on ventricular repolarization. ⁷ , ²¹ Moreover, pharmacological cardioversion in patients with already impaired ventricular repolarization is a class III recommendation (i.e., is contraindicated). ²² Among the many ion channels that are involved in cardiac repolarization, few offer the unique opportunity to primarily act on atrial electrophysiology, such as the SK‐channels.

4.2. Atrial antiarrhythmic properties and ventricular safety of SK‐channels‐inhibition in a pig model for obstructive respiratory events

Aside from being present in the central nervous system, where SK‐channels regulate neuronal excitability, these potassium channels have been found to be expressed in peripheral tissues, including the heart. In cardiac electrophysiology, SK‐channels contribute to phase 3 of repolarization when intracellular calcium peaks. ²³ Unlike most other channels governing the electrical activity in the heart, SK‐channels are not voltage‐gated, but instead open upon increased intracellular calcium concentrations. ²⁴ This has been demonstrated to be mediated by the Ca²⁺/calmodulin‐dependent protein kinase II (CaMKII), ²⁵ which previously has been reported to be enhanced in patients with sleep apnea. ²⁶ Specific pathological apnea‐related mechanisms, such as increased atrial stretch due to intrathoracic pressure swings, increased autonomic activation, and increase in reactive oxygen species due to repetitive hypoxemia, potentially independently contribute to enhanced activation of CaMKII in the setting of OSA. ²⁷ , ²⁸ , ²⁹ Additionally, atrial stretch is a potential contributor that can alter calcium sensitivity, CaMKII and therefore SK‐channel integrity. ³⁰ In our model, CVP was determined as a proxy‐parameter for right atrial load and stretch. INAP reproducibly increased CVP, which might contribute to atrial arrhythmogenesis in OSA. ³¹ , ³² Inhibition of SK‐channels could prevent stretch‐induced shortening of AERP in an explanted heart model, further underlining the importance of SK‐channels in a setting of increased atrial stretch. ³³

Pharmacological inhibition of SK‐channels in our model increased atrial refractoriness and could sufficiently blunt progressive shortening of atrial refractoriness and atrial action potential duration induced by INAP‐application, which protected against increased AF inducibility. However, increased atrial refractoriness was unlikely to contribute to the demonstrated low AF burden. In the same model, both sotalol and amiodarone prolonged AERP, but could not prevent a shortening in AERP related to INAP and could not blunt an increased INAP‐associated AF risk. ¹⁰ Since the model could not provide further mechanistic insights into whether SK‐channels are associated with OSA‐related AF, the atrial antiarrhythmic properties of SK‐channel inhibition, as demonstrated in this porcine model, warrant further investigation. Some results indicate an involvement of other channels aside from SK‐channels, which may be caused indirectly by SK‐channel inhibition. Postrepolarization refractoriness as described in this model is normally associated with sodium channel inhibition or functional class I blocking properties. Even though previous studies could not substantiate direct inhibitory effects of AP14145 on sodium channels, it has been suggested that SK‐channel inhibition may affect sodium current indirectly. ³⁴ Direct SK‐channel inhibition shifts the resting membrane potential toward more depolarized potentials, which may affect sodium current through accumulation of state‐dependent inactivated channels. ³⁵ Additionally, p‐wave duration (Supporting Information: Figure 2) was prolonged in the presence of AP14145, potentially indicating a slowed atrial conduction, which is an established consequence of class I blocking properties. It remains unclear though, the extent to which these effects contributed in blunting arrhythmogenic effects during INAP.

Protective effects against INAP‐associated arrhythmogenic effects due to pharmacological SK‐channel inhibition occurred without impairing ventricular repolarization. An impact on ventricular electrophysiology poses a problem when treating with AAD, but can also be problematic with OSA. ⁷ , ²² We previously described transient changes in QT‐interval duration and shortening of the EMW being associated with increased ventricular arrhythmia risk due to INAP, which was potentiated by hERG1‐blocking AAD, dofetilide. ⁷ The SK‐channel blocker investigated in this study, AP14145, was not associated with potentiated impairment of ventricular arrhythmogenic risk during obstructive respiratory events, which resembles potential drug safety, especially in a disease state, in which ventricular SK‐channel activity was previously suspected to be involved in. ³⁶ Heart failure is a common comorbidity in patients with AF and OSA and should be acknowledged in future drug safety assessment studies of SK‐channel‐inhibition. In both human and canine cellular studies, heart failure‐induced prolongation of ventricular repolarization was further prolonged by SK‐channel inhibition. ³⁷ In heart failure, SK‐channels may therefore be important for the stability of ventricular repolarization and inhibition of SK‐channels, for the treatment of AF in heart failure patients, may result in increased ventricular arrhythmogenesis. Thus SK‐channel inhibition in heart failure warrants further safety evaluation. ³⁷

Whether SK‐channel inhibitors can sufficiently treat AF‐patients with OSA remains to be further investigated in chronic OSA models and clinical trials. In this study, AP14145 was administered intravenously. An available oral compound is likely required before a feasible application in the clinic can be considered.

4.2.1. Study limitations

Our study depicted antiarrhythmic effects of SK‐channel inhibition in the setting of obstructive respiratory events. However, this does neither elucidate the actual underlying arrhrythmogenic mechanisms and does not prove an involvement of SK‐channels in OSA pathology. Blunted arrhythmogenic effects could have resulted due to potent prolongation of atrial refractoriness, which might overshadow underlying pathophysiological mechanisms. In our model, SK‐channel inhibition resulted in blunted apnea‐associated arrhythmia risk, where other AADs failed to do so. Unfortunately, due to the very transient and reversible nature of arrhythmogenic consequences to INAP, an extraction of cardiac tissue to analyze specific pathways was not feasible. Hence, we can only discuss putative mechanisms underlying this model.

OSA is a recognized chronic disease. Our model for OSA however did not cover chronic atrial remodeling, in which atrial fibrosis and enlargement might have further‐reaching implications for the antiarrhythmogenic potential for SK‐channel inhibition. Nevertheless, our model does provide the opportunity to focus on individual obstructive respiratory events, which allowed us to investigate arrhtyhmogenic effects of acute single obstructive apneas.

Additionally, expression levels of SK‐channels may vary depending on relevant comorbidities and the disease stage of AF. ³⁸ , ³⁹ In atrial tissue analysis of AF‐patients, certain SK‐channels (SK2‐isoform) have been described to be downregulated in the atria. ³⁹ Additionally, in cardiomyocytes of heart failure animal models and patients, certain subtypes of SK‐channels have been found to be increasingly expressed in ventricular tissue. ³⁸ This may have consequences for the atrial antiarrhythmic effect of SK‐channel inhibitors in patients with heart failure and may be an important limitation for the suggested atria‐specific profile of SK‐channel inhibition. Therefore, the role of SK‐channels as treatment targets for AF has to be further evaluated in OSA and with special regard to heart failure.

5. CONCLUSIONS

Pharmacological inhibition of SK‐channels prolongs atrial refractoriness and blunts arrhythmogenic changes induced by obstructive respiratory events in pigs. Whether SK‐channel inhibition can be a useful approach to treat AF patients with OSA requires further studies.

AUTHOR CONTRIBUTIONS

Benedikt Linz, Eva Melis Hesselkilde, Mark Alexander Skarsfeldt, Jonas Goldin Diness, Bo Hjorth Bentzen, Dominik Linz, and Thomas Jespersen conceived and designed research. Benedikt Linz, Mark Alexander Skarsfeldt, Eva Melis Hesselkilde, and Yannan Yan conducted animal experiments. Benedikt Linz and Julie Norup Hertel analyzed in vivo data. Benedikt Linz wrote the manuscript. Thomas Jespersen, Jonas Goldin Diness, Stefan Michael Sattler, Dominik Linz, and Jacob Tfelt‐Hansen helped in designing, consulting and project progression. All authors critically reviewed the manuscript.

Supporting information

Supporting information.

Linz B, Hesselkilde EM, Skarsfeldt MA, et al. Pharmacological inhibition of SK‐channels with AP14145 prevents atrial arrhythmogenic changes in a porcine model for obstructive respiratory events. J Cardiovasc Electrophysiol. 2023;34:126‐134. 10.1111/jce.15769

DATA AVAILABILITY STATEMENT

The data underlying this article are available in the article itself and in its online supplementary material.