Accelerated Sinus Rhythm Prevents Catecholaminergic Polymorphic Ventricular Tachycardia in Mice and in Patients

Animal studies

Animal use was in accordance with the National Institutes of Health guidelines and approved by the institutional animal care and use committee. Casq2−/− and RyR2^R4496C+/− mice generation and in vivo characterization have been previously described.^{10, 21}

Atropine treatment prior to ISO challenge and surface ECG recording

Experiments were carried out in a randomised crossover design. Surface ECGs were recorded from Casq2−/− mice (n=15) and wild-type (WT) littermates (n=11), challenged with 3mg/kg of isoproterenol (ISO) 8 minutes after injection of either vehicle (distilled water) or atropine (0.5mg/Kg). Arrhythmia incidence and arrhythmia score were determined in the atropine/ISO group and compared to the incidence and score in the same mice when treated with vehicle/ISO.

Atrial pacing in the presence of ISO challenge

Supraventricular overdrive stimulation was performed in anesthetized Casq2−/− (N=19) and RyR2^R4496C+/− mice (n=9) using transesophageal atrial pacing. Atrial stimulation rates were selected to exceed the maximum intrinsic sinus rate achieved during catecholamine challenge (i.e., 600 b/min for Casq2−/− mice and 660 b/min for RyR2^R4496C+/− mice). The atrial overdrive pacing was maintained for the entire period the CPVT mice develop ventricular ectopy in response to the catecholamine challenge. Arrhythmia incidence and arrhythmia score were compared in the same mice challenged with ISO/caffeine in the presence or absence of atrial pacing.

Single myocyte studies

Ventricular myocytes were isolated from Casq2−/− mice and loaded with Fura-2 AM for intracellular Ca measurements as previously described.¹⁰ Myocytes were field-stimulated at increasing pacing rates (0.5 – 4Hz) after adding ISO (1µM) to induce SCW. Ca fluorescence ratio (F_ratio) was recorded for 30s at each pacing rate and incidences of SCW and triggered beats quantified.

CPVT patient data

To evaluate the HR dependence of CPVT in humans, we screened the entire registry of RyR2-mutation positive CPVT patients (N=162) of the Academic Medical Center (Amsterdam, the Netherlands) for exercise tests that were done before antiarrhythmic drug therapy was started. A total of 42 drug-naïve patients with VA were identified. Exercise testing was performed on a treadmill (standard or modified Bruce protocol) or a bicycle ergometer with a gradually increasing workload protocol. Patients exercised until exhaustion. In order to test our hypothesis that a fast supraventricular rhythm can overdrive ventricular arrhythmias, only exercise tests from patients who reached at least 85% of their estimated maximum HR were used for the analysis (21 out of 42 patients tested). Of these 21 patients, 3 developed ventricular arrhythmias only at peak exercise and were excluded because the stress test was stopped before the suppression of arrhythmias could potentially occur by achieving a higher sinus rate. Thus, a total of 18 patients exercised beyond the onset of ventricular arrhythmias and were included in the analysis.

Statistical analysis

Data are presented as mean ± SEM. Arrhythmia scores and number of VPBs recorded were compared by means of the Mann-Whitney U test. Incidence of VA in the different groups was compared by Fisher exact test. A 2-tailed p value <0.05 was considered statistically significant.

Vagolytic therapy reduces the incidence of ventricular arrhythmias in a CPVT mouse model

Injection of atropine i.p. induced a 15% increase of the intrinsic sinus rate in Casq2−/− anaesthetized mice compared to control injection of vehicle in the same mice (atropine 430±7 bpm vs. vehicle 374±16 bpm, n=15, p<0.05, Fig. 1A). Eight minutes after atropine or vehicle injection, a catecholaminergic challenge with ISO was performed. As shown in Fig. 1B, ISO induced a robust HR increase in both groups. Peak heart rate was higher in the atropine/ISO group (604±12 in atropine/ISO vs. 565±19 in vehicle/ISO, n=15, p=0.09), whereas the magnitude of heart rate increase was comparable to that of the vehicle/ISO group.

Atropine pretreatment significantly reduced the ventricular arrhythmia score compared to vehicle/ISO controls (0.66±0.21 in atropine/ISO vs. 1.73±0.34 in vehicle/ISO, p<0.05, Fig. 2A). Of the 11 mice with significant ventricular ectopy during vehicle/ISO treatment, 8 (73%) showed an improvement of their ventricular arrhythmia score, with 5 mice having complete ventricular arrhythmia suppression. Of the 7 mice with ISO–induced VT, 6 (86%) were free of VT after pre-treatment with atropine (Fig. 2B). Only 2 out of 15 mice tested had a worsening of their ventricular arrhythmia score from 0 (no/sporadic VPBs) to 1 (frequent/bigeminal VPBs) with atropine/ISO (Fig. 2B).

We then quantified the ventricular arrhythmia burden (VPBs/mouse). Mice pretreated with atropine displayed a reduction of ISO-induced VPBs compared to vehicle-treated mice (20±7 in atropine/ISO vs. 51±17 in vehicle/ISO, N=15, Fig. 1B). Furthermore, the incidence of VPBs peaked before the peak of sinus HR was reached (Fig. 1B). In other words, VPBs were suppressed as the sinus HR rose further (Fig. 1B). Interestingly, the same dose of atropine (0.5 mg/kg) did not significantly alter the sinus HR in anaesthetized WT mice (427±10 bpm at baseline vs. 432±16 bpm 8 minutes after atropine injection, p=0.7). Together with the observation that Casq2−/− mice displayed a slower basal sinus HR compared to WT mice (388±8 bpm in Casq2−/− vs. 427±10 bpm in WT, p<0.05), this result indicates that vagal stimulation contributed to the sinus bradycardia of Casq2−/− mice.

Atrial pacing prevents ventricular arrhythmias in Casq2−/− mice

To control for possible confounding effects by atropine (e.g., direct pharmacological effects in the specialized conduction system and/or the ventricular myocardium), we next tested the effect of increasing supraventricular HR by transesophageal overdrive pacing in Casq2−/− mice. Atria were paced at 10Hz (600 bpm) to match the peak sinus rates obtained with a catecholaminergic challenge. Electrical capture of the atrium and appropriate AV node conduction was verified on ECG (Fig. 3A). In the absence of catecholamine challenge, rapid atrial pacing (10 Hz, 30 s) induced ventricular ectopy in only 1 out of 19 Casq2−/− mice tested. Intermittent second degree AV block was observed in 3/19 mice during initial control pacing.

In response to ISO challenge, 11/19 (58%) Casq2−/− mice developed ventricular ectopy (Fig. 3B). Atrial overdrive pacing completely prevented ventricular arrhythmias in 8 out of those 11 mice (73%), with only 3/19 (15%) mice overall developing ventricular arrhythmias (p<0.05, Fig. 3B). The ventricular arrhythmia score in the atrial pacing/ISO group was significantly reduced compared to ISO only control group (from 1.42±0.28 in ISO to 0.47±0.25 in atrial pacing/ISO, p<0.05, Fig. 3C). Overall, atrial overdrive pacing resulted in a 10-fold reduction in the number of ISO-induced VPBs (from 58±16 to 7±4 VPB/mouse, p<0.01, Fig. 3D).

Prolonged atrial pacing prevents ventricular arrhythmias in RyR2^R4496C+/− mice

Since heterozygous mutations in the RyR2 gene are the most common cause of CPVT in patients, we next tested our hypothesis in a RyR2-linked CPVT mouse model, RyR2^R4496C+/− mice. ²¹ Compared to Casq2−/−, RyR2^R4496C+/− mice displayed both a higher heart rate at baseline (384±13 in Casq2−/− vs. 472±12 in RyR2^R4496C+/−, p<0.001), and a higher peak sinus rate after ISO challenge (555±17 in Casq2−/− vs. 653±10 in RyR2^R4496C+/−, p<0.001). Unlike Casq2−/− mice, RyR2^R4496C+/− did not display ventricular ectopy in response to ISO only (3 VPBs in 1/7 mice tested),. Hence, we used a combination of caffeine (120 mg/kg) and ISO (3mg/kg) as catecholamine challenge to induce arrhythmias.

Injection of ISO/caffeine induced a rapid HR increase in all mice (n=9) and ventricular arrhythmias in 66% of them (Fig. 4A–B). However, in contrast to Casq2−/− mice, where ectopy appeared during the first few seconds of the HR upstroke, RyR2^R4496C+/− developed ventricular arrhythmias on average 3.5±0.8 min after injection and the episodes lasted longer (7.1 ±0.8 min until ventricular arrhythmias stopped). Interestingly, at the onset of ventricular ectopy the HR in RyR2^R4496C+/− mice had already slowed down to 68.5±6.7% of the maximum HR reached (Fig. 4A). Prolonged overdriving pacing (660bpm for 10min) successfully prevented ventricular arrhythmias in all but one mouse and reduced the ventricular arrhythmia score (from 1.8±0.58 in ISO/caffeine to 0.11±0.11 in atrial pacing-ISO/caffeine, p<0.05) and the number of VPBs recorded (from 543±239 VPBs in ISO/caffeine to 0.55±0.55 VPBs in atrial pacing-ISO/caffeine, p<0.05) compared to the same mice treated with ISO/caffeine in the absence of atrial pacing (Fig. 4B–C–D).

Overdrive pacing supresses ISO-induced SCWs in Casq2−/− myocytes

To further investigate the underlying mechanism of arrhythmia suppression at fast heart rates, we next tested the effects of progressively shorter diastolic interval in single myocytes of Casq2−/− mice. In presence of ISO (1µM), myocytes were field-stimulated at 0.5, 1, 2, 3, and 4 Hz. SCWs were present in approximately 40% of Casq2−/− paced at 0.5 and 1 Hz. Reduction of the diastolic interval by higher pacing frequencies progressively reduced the incidence of SCWs (Fig. 5). At 3Hz only 8% of Casq2−/− myocytes developed SCWs and complete prevention of SCWs was achieved with overdriving pacing at 4Hz (Fig. 5). Taken together, consistent with our in vivo experiment, overdrive pacing prevents ISO-induced SCWs in CPVT myocytes.

Sinus tachycardia paradoxically suppresses arrhythmias in a subpopulation of CPVT patients

The exercise tests of 18 CPVT patients carrying a mutation in the RyR2 gene met our inclusion criteria and were therefore used for the analysis. Basal HR was 76±4 bpm. As expected, at the beginning of the stress test patients did not develop ventricular ectopy. With increasing exercise, all 18 patients exhibited ventricular ectopy. Three of them displayed couplets or VT as their first ventricular arrhythmia, the other 15 patients had frequent VPBs and bigeminal VPBs first. In 86% of the latter group, ventricular arrhythmia complexity increased leading to couplets (13/15, 86%) and VT (6/15, 43%). HR at the onset of ectopic activity was variable ranging from 95 to 150 bpm (mean 128±4 bpm). The first ventricular arrhythmia appeared at 65% of peak HR as shown in Fig. 6A. Upon reaching 87% of their maximum HR peak, all the patients had developed ventricular arrhythmias. However, in 6 out of 18 patients (33%), ventricular arrhythmias subsided as their sinus HR further increased up to 89%, 91%, 93% (in 2 patients), 95% and 99% of their peak HR (Fig. 6A). The most severe arrhythmic phenotype displayed in these 6 patients before sinus tachycardia took over was bigeminal VPBs in one patient, couplets in 4, and VT in one (Fig. 6B). Taken together, we find VA suppression with increasing sinus tachycardia in a subset of CPVT patients. We next compared the HR response in “responders” (=VA suppression by exercise) and “non-responders” (=No VA suppression). As shown in table 1, resting HR, HR at the beginning of arrhythmia and HR increase during exercise were not significantly different in the two groups. There was a trend that responders exercised for a longer period of time and at a higher intensity (MET values) compared to non-responders.

Main findings

Sinus bradycardia^{1, 2, 9, 10, 12, 14, 15} and sinatrial node dysfunction^{16, 17, 22, 23} can be found in CPVT patients and CPVT mouse models. Here, we tested the hypothesis that augmenting supraventricular heart rates prevents CPVT in vivo. We find that both vagolytic therapy (Figs. 1 & 2) and atrial overdrive pacing (Fig. 3 & 4) can suppress catecholamine-induced ventricular arrhythmia in Casq2−/− and RyR2^R4496C+/− mice. Our in vitro studies confirm that reduction of diastolic interval suppresses ISO induced-SCW in isolated Casq2−/− myocytes, thereby reducing the risk of DADs and triggered activity (Fig. 5). In addition, we observed that in a subset of CPVT patients, exercise-induced ventricular arrhythmias were paradoxically suppressed as sinus heart rates increased further with continued exercise (Fig. 6).

Antiarrhythmic effect of vagolytic treatment in a CPVT mouse model

Pretreatment with atropine, which blocks muscarinic receptors²⁴ and increases SA node firing rates,²⁵ not only significantly reduced the overall ventricular arrhythmia score and the number of VPBs compared to controls, but also prevented the most severe triggered events (couplets and VTs). We hypothesize that the atropine-induced acceleration of the supraventricular pacemaker caused an upward shift of the entire HR response curve to ISO (Fig. 1). As a result, the diastolic intervals in the ventricular tissue is shorter at every heart rate, thereby reducing the likelihood that spontaneous Ca waves and DADs reach threshold and triggered beats. It could be argued that during CPVT, mice can reach higher ventricular rates²⁶ than the fastest sinus rate obtained in atropine treated mice (Fig. 1). However, when ventricular arrhythmias start, usually with sporadic VPBs and bigeminy, the HR increase is still moderate (Fig. 1). It is only as ventricular arrhythmia complexity worsens that the coupling time of ectopic beats progressively reduces until fast VT is triggered. We hypothesize that atropine induced accelerated sinus rhythm may be sufficient to interrupt simple arrhythmias such as sporadic VPBs and bigeminy, arresting the likely progression of these events into ventricular tachycardia.

Moreover, it will take at least a few seconds for the ISO-induced signal transduction events to occur that promote SR Ca overload and triggered arrhythmias.²⁷ By the time these effects have taken place, vehicle treated mice develop complex arrhythmias (8 seconds after ISO injection), whereas atropine treated mice are protected possibly due to their higher HR and hence shorter diastolic interval. Our in vitro studies (Fig. 5) confirm that a reduction of diastolic interval suppresses ISO induced-SCR in isolated Casq2−/− myocytes, thereby reducing the risk of DADs and triggered activity. We speculate that the reduction of diastolic interval might be particularly effective in the Purkinje cells (PC) of the conduction system. In fact, it has recently been shown that PC are more prone to developing DADs than working ventricular myocytes²⁸ because of intracellular characteristics that increase the Ca handling dysregulation in these cells.²⁹

The data from the RyR2^R4496C+/− CPVT mouse model reported here also support our hypothesis: compared to Casq2−/− mice, RyR2^R4496C+/− mice have significantly higher resting heart rates, and a significantly bigger response to ISO (compare Fig. 3 and 4), similar to Casq2−/− mice pretreated with atropine, which are resistant to ISO challenge (Fig. 1). In RyR2^R4496C+/−, ventricular arrhythmias can only be induced after sensitizing the RyR2 channels with caffeine, and even then the ventricular ectopy occurs only when the intrinsic sinus rate drops below 70% of its peak value (Fig. 4). Other groups have also reported that unlike Casq2−/− mice, RyR2^R4496C+/− do not exhibit spontaneous or exercise-induced ventricular arrhythmias¹⁵. Pharmacological challenges that include the use of catecholamines and/or caffeine are always required to elicit arrhythmias in this model.^{15, 30}. Our data demonstrate that overdrive pacing was effective also against drug-induced ventricular arrhythmias in RyR2^R4496C+/− mice.

Interestingly, atropine injection increased the sinus HR to a much greater extent in Casq2−/− compared to WT mice. This result suggests that vagal tone contributes to the sinus bradycardia observed in Casq2−/− mice, either due to increased vagal tone, or due to increased responsiveness of the SA node to vagal stimulation. Since sinus bradycardia^{1, 2, 12} and SA node dysfunction^{16, 17, 22} are also observed in CPVT patients, it is conceivable that an impaired sinus rate response during exercise contributes to the exercise-induced ventricular ectopy in CPVT patients. Consistent with this hypothesis, we observed complete ventricular arrhythmia suppression as the HR further increased approaching its peak in a subgroup of patients with significant ventricular ectopy in the early stages of the exercise (Fig. 5). This occurred despite the fact that catecholamine-induced HR increase initially is accompanied by progression of ventricular arrhythmia complexity.² Although it is well recognized that ectopic ventricular automaticity can be reset by overdrive pacing,^31–33 it is still unexpected that sinus tachycardia could provide sufficient protection against clinical CPVT. A possible explanation is that the positive chronotropic effect of β adrenergic stimulation during stress could paradoxically counterbalance the triggered activity simultaneously promoted by β-adrenergically induced SR Ca overload. However, possibly due to the small number of patients and the heterogeneity of exercise test protocols used, we could not detect a significant difference in the rate of HR increase in the subset of patients with arrhythmia suppression compared to the rest (table 1). A prospective study on a larger population would be needed to clarify the relationship between HR increase and arrhythmia suppression in CPVT patients.

Antiarrhythmic effect of atrial overdrive pacing in CPVT

As important proof of principle, we find that atrial overdrive pacing almost completely prevented catecholamine-induced ventricular arrhythmia in both Casq2−/− and RyR2^R4496C+/− mice (Fig. 3 & 4). Finding alternative treatment approaches is an important issue for CPVT patients, since β-blockers, which are the current gold standard in CPVT treatment, are not fully protective in up to 30% of the patients.^{34, 35} Based on our experimental results reported here, one might speculate that the negative chronotropic effect of β-blockers might contribute to treatment failure. In fact, by slowing SA node activity, β-blockers may suppress an intrinsic antiarrhythmic mechanism of the heart rate increase. In support of this hypothesis, recent work has shown that while β-blockers protect most CPVT patients from severe exercise-induced arrhythmias, they reduce the sinus HR threshold at which milder ventricular arrhythmias start.³⁶ On the other hand, molecules that regulate the RyR2 open probability (such as flecainide, propafenone and RyR2 inhibitors S107, JTV519) reduce the frequency of diastolic SR Ca waves and triggered beats.^{13, 37–39} Consistent with this mechanism, flecainide increased the sinus HR threshold at which ventricular arrhythmias were triggered clinically, while at the same time reducing the peak HR reached during exercise.⁴⁰

Clinical implications and caveats

While vagolytic therapy would be difficult to carry out in CPVT patients because of systemic side effects,⁴¹ overdrive atrial pacing could be performed in patients with dual chamber pacemakers or ICDs. In support of this concept is a recent clinical case report of the temporary overdrive suppression of bidirectional VT by a rapid atrial tachycardia in a young CPVT patient.⁴² One concern is that increasing the HR will independently load the SR with Ca⁴³ and may promote the spontaneous Ca release events that trigger CPVT. We tested this hypothesis experimentally, and only found one instance where rapid atrial pacing induced ventricular ectopy. Our data in CPVT patients further demonstrates that at least 30% of CPVT tolerated very fast sinus rates with paradoxical suppression of ventricular arrhythmia. To what extent this finding is applicable to other CPVT patients remains to be determined. For example, the incidence of atrial tachyarrhythmia is increased in CPVT patients^{1, 17} and atrial fibrillation has been associated with triggering CPVT clinically.⁴⁴ Another important consideration is the site of overdrive pacing. Given that electrical stimulation by the pacing lead may trigger the release of norepinephrine from sympathetic nerve terminals in the ventricle,⁴⁵ which would further exacerbate the ventricular automaticity, atrial pacing likely will be the preferable approach. In either case, atrial overdrive pacing will have to be tested in humans with CPVT to evaluate its safety and efficacy. If rapid atrial pacing is shown to prevent exercise-induced VT in CPVT patients, it may be feasible to automatically program high atrial rates during exercise using the rate-response feature.⁴⁶