Urgent catheter ablation for sustained ventricular tachyarrhythmias in patients with acute heart failure decompensation

Abstract

Aims

Ventricular tachycardia (VT) and ventricular fibrillation (VF) are not uncommon in patients hospitalized with acute heart failure (AHF). We sought to evaluate the efficacy of urgent radiofrequency catheter ablation (RFCA) for recurrent VT/VF during AHF decompensations.

Methods and results

The present study retrospectively analysed the data of 15 consecutive patients (69 ± 9 years, ischaemic heart disease in 10), who underwent urgent RFCA for frequent drug-refractory VT/VF episodes during an AHF decompensation with pulmonary congestion. The target arrhythmias were clinically documented monomorphic VTs in 10 patients, frequent premature ventricular contractions (PVCs) triggering VF in 4, and both in 1. The mean left ventricular ejection fraction was 26 ± 8%. The maximum number of arrhythmia episodes over 24 h was 9.1 ± 11.7. All RFCA sessions were completed without any major complications except for a temporary deterioration of pulmonary congestion in three patients (20%). Elimination and non-inducibility of the target arrhythmias were achieved in 13 patients (87%). Successful ablation site electrograms showed Purkinje potentials for all 5 PVCs triggering VF and 4 of 14 clinically documented monomorphic VTs (29%). Five patients (33%) underwent second sessions 10 ± 4 days after the first session for acute recurrences. Sustained VT/VF was completely suppressed during admission in 12 patients (80%), and the AHF ameliorated in 13 patients (93%). Twelve patients (80%) were discharged alive.

Conclusion

Urgent RFCA for drug-resistant sustained ventricular tachyarrhythmias during AHF decompensations would be an appropriate therapeutic option. Purkinje fibres can be ablation targets not only in those with PVCs triggering VF, but also in those with monomorphic VT.

Introduction

Ventricular arrhythmias are commonly seen in the setting of acute heart failure (AHF). Treating frequent ventricular tachycardia (VT) or ventricular fibrillation (VF) episodes during an AHF decompensation is quite challenging. Antiarrhythmic drug administration is the first-line therapy, but Class I antiarrhythmic drugs often fail to maintain sinus rhythm and may worsen heart failure.¹ Amiodarone is more effective, but complete arrhythmia suppression is difficult, especially in those suffering from incessant tachyarrhythmias.^1,2 A sympathetic blockade was reported to be superior to antiarrhythmic therapy in those with electrical storms,² but the overall mortality was still high. Furthermore, beta-blockers can exacerbate heart failure if administered during an AHF decompensation.² Radiofrequency catheter ablation (RFCA) was shown to be effective in treating electrical storms,^3–5 but, to our knowledge, no ablation studies have enroled patients specifically in the setting of AHF. In the present study, we sought to evaluate the acute efficacy following clinical outcomes of urgent RFCA in a selected population of patients with sustained VT/VF during an AHF decompensation.

Methods

Study population

Over the past 10 years, 176 patients with clinically documented sustained ventricular arrhythmias underwent an electrophysiological study and RFCA at the Nippon Medical School Teaching Hospital. Among those, the present study included 15 consecutive patients (Patients 1–15 in Table 1, 11 men, age 69 ± 9 years) in whom tachyarrhythmias developed and an RFCA procedure was carried out during an AHF decompensation. The diagnosis of heart failure was based on the signs, symptoms, and imaging.⁶ All patients complained of breathlessness, presented with audible moist rales, and exhibited pulmonary congestion on chest X-ray. The targeted arrhythmias of the RFCA were all clinically documented sustained ventricular tachyarrhythmias, which were monomorphic VTs in 10 patients (66%), monomorphic premature ventricular contractions (PVCs) preceding VF in 4 (27%) (Figure 1), and both in 1 other (7%).

Electrophysiological study and catheter ablation

Written informed consent for the electrophysiological study and catheter ablation was obtained from the patients or legal representatives. Except for Patient 13, who underwent inhalational anaesthesia using sevoflurane, the study was performed under sedation using midazolam or propofol. The vital signs, respiratory status, fluid input, and urine output during the procedure were carefully monitored with direct arterial blood pressure monitoring, oxygen saturation monitoring, and a urinary tract catheter. Except for Patients 2 and 9, the administration of antiarrhythmic drugs and/or intravenous beta-blockers was continued: amiodarone, nifekalant, mexiletine, and landiolol in 12, 1, 2, and 2 patients, respectively.

Quadripolar electrode catheters were positioned at the His bundle region and right ventricle. Surface and bipolar endocardial electrograms were monitored continuously, and recorded on either an EPLab (Quinton Electrophysiology Corp.) or EP-WorkMate (EP MedSystems Inc.) recording system. In the earlier patients (Patients 1–8), mapping of the target arrhythmias was performed with a non-irrigated 4 mm or 8 mm tip deflectable quadripolar electrode ablation catheter (RF Marinr, Medtronic Inc., Blazer II, Boston Scientific Inc., and Navistar, Biosense Webster). After they became available in Japan, the ablation was carried out with 3.5 mm tip irrigated ablation catheters (Navistar Thermocool, Biosense Webster). An electroanatomical mapping system (CARTO, Biosense Webster) was used in Patients 7–15. The low-voltage areas and scar areas were defined by bipolar electrograms exhibiting an amplitude of ≤1.5 and ≤0.1 mV, respectively. Ablation was performed by the delivery of radiofrequency current between the distal electrode of the mapping catheter and a cutaneous adhesive electrode patch, using an appropriate radiofrequency generator for each catheter (Atakr RF Power Generator, Medtronic Inc., EPT1000TC, Boston Scientific Inc., or Stockert J70 RF Generator, Stockert GmbH). For the non-irrigated-tip catheter and irrigated-tip catheter, the radiofrequency energy delivery settings were a power of ≤50 and ≤35 W, respectively, and a temperature of ≤55 and ≤45°C, respectively.

In patients with monomorphic VTs, programmed right or left ventricular stimulation was carried out to induce the VTs, using a maximum of three extrastimuli at two different cycle lengths. If clinically documented VT was induced and haemodynamically tolerated, activation mapping with repeat entrainment pacing trains was performed. The target site of the monomorphic VT was identified if isolated diastolic electrograms showing the following criteria were observed: entrainment with concealed fusion, post-pacing interval nearly identical to the tachycardia cycle length, and the stimulus to QRS interval equal to the diastolic potential to QRS interval.⁷ During the mapping, attention was paid to try and detect any Purkinje potentials, because they are known to be included, at least in part, in the reentrant circuit of some VTs arising from a diseased heart.^8,9 In those demonstrating a documented VT with haemodynamic intolerance, substrate mapping identifying the isolated delayed potentials inside a low-voltage area during sinus rhythm was performed,¹⁰ and sequential RFCA lesions were created to transect all identified conducting channels where pacemapping exhibited a stimulus to QRS interval delay and a similar QRS morphology in the surface electrocardiographic leads as that during the VT.¹¹ In the five patients with PVCs preceding VF, radiofrequency energy applications were carried out at the earliest activation site of the PVCs where a similar paced QRS morphology as spontaneous PVC was recorded. After the ablation, the above-mentioned programmed electrical stimulation protocol was performed. No spontaneous arrhythmia occurrence and the non-inducibility of any clinically documented ventricular arrhythmias were defined as an acute success. In case of the induction of undocumented VTs by programmed stimulation, it depended on the operators whether they tried to ablate them or not. In those in which the initial RFCA failed or in those with recurrent tachyarrhythmias, repeat RFCA procedures were performed.

Statistical analysis

The data are expressed as the mean ± standard deviation for continuous variables, and as absolute frequencies and percentages for categorical variables. Differences between groups were not compared because of insufficient statistical power. Differences between the means of the tachyarrhythmia frequency before and after the RFCA, or between the means in VT cycle lengths for the first occurrence and during the RFCA session were assessed with Student paired t-test. The time to death after the RFCA was analysed with the use of a Kaplan–Meier curve and the estimated 12- and 24-month survival rates were presented. All tests were two-sided and a P < 0.05 was considered statistically significant. All statistical analyses were conducted using SPSS for Windows 11.0 J software (SPSS Inc.).

Results

Characteristics of the study subjects

During the 10-year study period, 1069 patients were admitted for AHF decompensations, and 15 patients were enroled into the present study, which accounted for 1.4% of those 1069 patients. Table 1 shows the clinical characteristics of these patients. They were admitted directly to our hospital for an AHF exacerbation (n = 6) or acute myocardial infarction (AMI) (n = 4), or were referred from other hospitals due to frequent monomorphic VT episodes that developed during treatment of the AHF (n = 5). Ten patients (67%) had coronary artery disease, three (20%) had non-ischaemic dilated cardiomyopathy, and two (13%) had a dilated phase of hypertrophic cardiomyopathy. The interval from the onset of the symptoms to admission was 2.3 ± 2.3 days in the 11 patients admitted for an AHF decompensation and from 2 to 36h in the 4 patients admitted for an AMI who also showed signs of AHF at the time of admission. Twelve patients (80%) had peripheral oedema, nine (60%) exhibited pleural effusion on chest X-ray, and six (40%) were in cardiogenic shock at the time of admission. All of them exhibited a depressed left ventricular systolic function with a mean left ventricular ejection fraction of 26 ± 8%. Eleven patients (73%) had been diagnosed with heart failure 6.1 ± 4.4 years before admission, and heart failure became exacerbated before admission. New-onset AHF was seen in another four, and all suffered from ischaemic heart disease including an AMI in three patients. Six patients (40%) had undergone an implantation of a defibrillator for sustained VT 3.8 ± 3.6 years before admission, and five patients (33%) subsequently experienced appropriate defibrillator therapies. An electrical storm was recorded in two of them, which was, then, suppressed by antiarrhythmic drugs.

Sustained VT/VF episodes requiring direct current deliveries occurred 5 ± 8 days after admission (median 7 days). Fifteen monomorphic VT morphologies were recorded in 11 patients, and frequent monomorphic PVCs triggering VF were observed in 5 patients (Figure 1). Ventricular fibrillation triggered by monomorphic PVCs was observed in 3 of the 4 patients (75%) with new-onset AHF, and in 2 of the 11 patients (18%) with worsening pre-existing heart failure. Class III antiarrhythmic drugs (amiodarone and/or nifekalant), Class Ib drugs (lidocaine or mexiletine), and intravenous beta-blockers were administered in 14 (93%), 8 (53%), and 11 patients (73%), respectively, but failed to prevent tachyarrhythmias (Table 1). An electrical storm (≥3 separate attacks within 24h) was recorded in 12 (80%). The maximum incidence of sustained VT/VF attacks within 24h was 9.1 ± 11.7, and an average daily frequency from the day of initial VT/VF occurrence to that of the RFCA was 2.7 ± 2.3. After occurrence of the tachyarrhythmias, coronary angiography was performed in all patients with coronary artery disease, and percutaneous coronary intervention was carried out in five patients showing residual coronary stenosis, resulting in the absence of a stenosis of >50% in any coronary vessel. Sustained tachyarrhythmias, however, were not suppressed, and we decided to carry out an urgent RFCA 6 ± 4 days after the first occurrence of the arrhythmias, which was 11 ± 8 days after admission. Before and during the RFCA procedure, intravenous catecholamines, mechanical ventilation, intra-aortic balloon counterpulsation, and continuous haemodiafiltration were used according to the patients' status (Table 1), but moist rales were audible and pulmonary congestion was visible on chest X-ray in all study subjects.

Electrophysiological findings and acute results of radiofrequency catheter ablation

In all patients, the RFCA sessions were completed without an interruption due to further AHF deterioration, respiratory failure, or other complications. Except for one VT in Patient 6, all clinically documented monomorphic VTs were induced by programmed stimulation. The cycle lengths of the 14 documented VTs were significantly longer during the ablation session compared with the initially documented cycle lengths on the ward (430 ± 35 vs. 382 ± 40 ms, P < 0.001). The aetiologies of these VTs were reentry in 13 (93%) and a non-reentrant focal mechanism in 1 other (7%) (Table 2). Among the 13 reentrant VTs, entrainment mapping could be performed in 11 VTs, and was not possible due to haemodynamic deterioration in the other 2, in which substrate mapping was carried out. The successful ablation site or estimated VT origin is illustrated in Figure 2.

In 4 of the 14 monomorphic VTs (29%), Purkinje fibres were involved in the origin of the tachycardia (Figure 2). Patients 2 and 3 demonstrated a reentrant VT with right bundle branch block and a superior axis morphology, which originated from the left posterior Purkinje fibres. The mechanism was analogous to idiopathic left VT; a presystolic Purkinje potential preceding the onset of the QRS and His bundle electrogram was seen and recorded earlier in the distal portion of the left posterior septum during the VT. The detailed VT mechanism has been previously described in another report.⁹ Patient 10 exhibited both a drug-refractory monomorphic regular VT and PVCs triggering VF (Figures 3 and 4). The mechanism of the VT was considered non-reentrant focal firing as it was not induced, reset, or entrained by programmed stimulation. During the VT, the His bundle electrogram preceded the onset of the QRS and an identical pacemap was obtained at the site showing a left anterior Purkinje potential during sinus rhythm and VT (Figure 3). Patient 11 had two incessant monomorphic VTs and one of them was bundle-branch reentry.⁸ The average QRS width of these Purkinje-related VTs was 155 ± 26 ms, and that of the other 10 VTs arising from the myocardium was 202 ± 24 ms. In the first session, acute success was achieved in 12 of the 14 VTs targeted (86%), including all Purkinje-related VTs. In Patients 6 and 11, one VT each was still inducible at the end of the RFCA session (Figure 2). In these two VTs, the duration from the QRS onset to the earliest rapid deflection in the precordial leads was 40 and 78 ms, and that from the QRS onset to the peak of the R-wave in V2 was 119 and 118 ms, respectively, suggesting an epicardial VT origin.¹² An epicardial approach, however, was not carried out in any of the present study subjects.

During the RFCA targeting the monomorphic PVCs triggering VF, the documented PVCs frequently occurred at least once every 2min until successful radiofrequency energy delivery in all patients. The site of the PVC origin was identified at the left posterior fascicle in four patients and left anterior fascicle in the remaining one (Figures 2 and 4). At those sites, Purkinje potentials preceded the onset of the QRS by 27 ± 10 ms during sinus rhythm and 51 ± 10 ms during the PVC, and ventricular pacing exhibited similar QRS morphologies to the PVCs in 10.2 ± 1.8 of the 12 body surface electrocardiographic leads. After the RFCA, the PVCs were suppressed in all patients.

Induction of an undocumented VT was observed in eight study subjects (53%), and an RFCA of those VTs was attempted in four patients, resulting in the non-inducibility of any VTs in two of them. The procedure time from the first puncture to removal of the sheaths was 179 ± 83min, and total radiofrequency energy delivery time was 785 ± 615s. The mean number of radiofrequency lesions was 17.7 ± 9.63. In Patients 9–15 who underwent the RFCA with an irrigated ablation catheter, the total amount of infused saline through the catheter during the session was 498 ± 340 mL.

Short-term follow-up results during admission

No procedure-related major complications were observed except for transient exacerbations of pulmonary congestion after the session in three patients (20%; Patients 5, 8, and 10). Among those three, an irrigated ablation catheter was used in only one (Patient 10) with a saline infusion of 298 mL. The pulmonary congestion was ameliorated with an increased dose of diuretics in three and catecholamines in one. Among the patients, except for those undergoing continuous haemodiafiltration, acute kidney injury, defined as an absolute increase in serum creatinine concentration of ≥0.3 mg/dL within 48h, occurred in Patients 5 and 6, but it ameliorated 2 and 3 days after the session, respectively, without any specific treatment. Among seven patients mechanically ventilated before and during the RFCA (Table 1), weaning was achieved in six patients 6 ± 4 days after the RFCA. The remaining patient (Patient 14) underwent a tracheotomy and received continuous positive airway pressure treatment throughout admission until death. Intra-aortic balloon counterpulsation was stopped in all five patients 3 ± 2 days after the RFCA. Disappearance of pulmonary congestion was confirmed in 14 patients (93%) 6 ± 3 days after the session.

Monomorphic VT recurred in five patients (33%; Patients 6, 8, 10, 11, and 13). The average daily frequency of the sustained VTs in these patients decreased compared with that before the RFCA (0.5 ± 0.3 vs. 2.4 ± 1.3 times a day, P = 0.04), and no electrical storms were observed. The morphology of the VT was similar to the previously targeted VT in four and different in one (Patient 11). A second session was carried out 10 ± 4 days after the first session, and non-inducibility of the documented VTs was achieved in four of the five patients (Patients 6, 10, 11, and 13). Even after the second RFCA session, Patients 6, 8, and 11, experienced a monomorphic VT recurrence 5, 9, and 2 days later, respectively. The average daily frequencies of sustained VTs, however, further decreased and were significantly reduced compared with that before the first RFCA session (0.2 ± 0.1 vs. 3.1 ± 1.0 times a day, P = 0.03).

Among the 15 study subjects, 12 patients (80%) including the 3 without complete suppression of the sustained VTs were discharged alive. An implantable defibrillator was used in all patients and oral amiodarone was prescribed in 11 (92%). Patients 2, 7, and 14 died in hospital due to an in-stent thrombosis, gastrointestinal bleeding, and low cardiac output syndrome following a surgical left ventriculoplasty 18, 10, and 39 days after the RFCA procedure, respectively. No sustained ventricular arrhythmias developed after the RFCA in any of these three patients.

Clinical outcomes after discharge

During the follow-up period of 33 ± 22 months, 5 (Patients 3, 6, 8, 11, and 13) of the 12 patients (42%) who were discharged alive experienced appropriate therapies from the implantable defibrillator for sustained VTs. The average interval from hospital discharge to the occurrence of arrhythmia was 3 ± 5 months. Five patients died 37 ± 27 months after hospital discharge: due to heart failure in Patients 3, 4, 6, and 11, and amiodarone liver toxicity in Patient 1. There was no sudden cardiac death. The estimated survival rate of all 15 patients at 12 and 24 months after the RFCA was 73 and 65%, respectively.

Discussion

RFCA during acute heart failure decompensation

To the best of our knowledge, there have been few studies which specifically included patients undergoing RFCA for ventricular arrhythmias during decompensated AHF. Several reports,^3,4 showing the efficacy of catheter ablation for electrical storms, included some patients with AHF which might have resulted from recurrent tachyarrhythmias. In our study subjects, all sustained ventricular arrhythmias developed during treatment of AHF, indicating that the VT/VF was not a cause but a consequence of the AHF. The aetiology of frequent arrhythmias during an AHF decompensation remains elusive, and probably is multifactorial, including an increased sympathetic tone, electrolyte imbalance, and myocardial stretch.¹³ We consider that new-onset or worsening heart failure forms a subentity of electrical storms. One study¹⁴ showed that heart failure was a precipitating factor of an electrical storm in 9% of the patients experiencing this life-threatening syndrome.

The present study included patients with a severe AHF decompensation; however, none of the sessions were interrupted due to haemodynamic intolerance, respiratory failure, or other complications. The procedures were completed with ordinary equipment, and no complications occurred except for transient pulmonary congestion. Acute success was achieved in 87% of the study subjects, and even in those without complete suppression of the tachyarrhythmias, the frequency significantly decreased. Carbucicchio et al.⁴ also reported suppression of the electrical storm during the short-term period after the RFCA even in those with residual inducibility of clinically documented VT. These results indicate that an urgent ablation is safe and would be an optimal therapeutic option in those with drug-resistant ventricular arrhythmias during an AHF decompensation.

It is unclear whether the amelioration of AHF was accelerated by elimination or decrease in the frequency of the tachyarrhythmias after the RFCA, but previous studies reported adverse tissue and cardiac mechanical responses to large electric currents,¹⁵ which were frequently delivered in all of the present study subjects before the RFCA. Besides the direct harmful effect of a high-energy current, the frequent occurrence of tachyarrhythmias might also have been an important exaggerating factor of myocardial stress.¹⁶

Factors related to the success of urgent RFCA

In the present study, an urgent RFCA was successfully performed in most of the patients with only an endocardial approach and lower upper limit of the ablation power settings than the other studies that performed VT ablation.^4,17 It may be due to the characteristics of the patients and arrhythmias. Ten patients had coronary artery disease, in which myocardial damage and arrhythmic substrates mainly existed on a relatively endocardial side, and eight patients exhibited ventricular arrhythmias involving Purkinje fibres, which are located just beneath the endocardium.¹⁸ On the other hand, two patients without acute success in the initial RFCA procedure suffered from non-ischaemic dilated cardiomyopathy, which is reported to be an independent predictor of failure of the RFCA procedure in patients with electrical storms,⁴ and the body surface electrocardiogram of their VTs suggested epicardial origins.¹² Epicardial VT ablation was not performed in any of the present study subjects, because it was considered more invasive than an endocardial approach alone. To further increase the rate of a successful ablation, however, an epicardial approach may be appropriate in those with signs suggesting an epicardial VT origin.

A continuous antiarrhythmic drug therapy might be another contributing factor. Presumably, due to drug efficacy, the clinically documented VTs induced during the session exhibited significantly longer cycle lengths than those which developed on the ward, and such longer cycle lengths might contribute to the haemodynamic stability and successful entrainment mapping achieved in 11 of the 13 patients with reentrant VT.¹⁹

Purkinje-related tachyarrhythmias in ischaemic heart disease

Purkinje fibres are shown to remain almost structurally preserved and physiologically viable even after an extensive myocardial infarction.^18,20 Such surviving Purkinje fibres exhibited a decreased resting membrane potential, maximum depolarization velocity and action potential amplitude, and increased action potential duration,²⁰ leading to spontaneous diastolic depolarization or reentry within the subendocardial Purkinje network.²¹

Regarding the life-threatening tachyarrhythmias arising from Purkinje fibres, Bänsch et al.³ reported four AMI cases with polymorphic VT or VF preceded by PVCs from Purkinje fibres. In the present study, Purkinje-related PVCs triggering VF were also seen in five patients with ischaemic heart disease including three with an AMI. Furthermore, Purkinje fibres were involved in the arrhythmia substrate in about 30% of the monomorphic VTs observed during AHF decompensation. Except for the bundle-branch reentry seen in one with non-ischaemic dilated cardiomyopathy, these Purkinje-related VTs were also observed in patients with ischaemic heart disease. The Purkinje-related VTs had a relatively narrow QRS width, and all were successfully treated by RFCA. These results suggest that, in those with ischaemic heart disease who suffered from frequent drug-refractory PVCs preceding VF or monomorphic VTs with a narrow QRS width, there should be no hesitation to perform RFCA even in the setting of an AHF decompensation.

Limitations

In the present study, the data were analysed retrospectively and there was no control group. During the study period, there might also have been patients suffering from AHF and drug-resistant lethal tachyarrhythmias, who underwent other therapeutic strategies, such as percutaneous cardiopulmonary support. Further study with a control group is warranted to clarify the actual benefit of RFCA in this category of patients. Also, we cannot affirm the optimal timing of the RFCA. We performed the ablation 6 ± 4 days after the first occurrence of these potentially lethal arrhythmias, but an earlier ablation procedure might have increased patient survival. Recently, Deneke et al.⁵ reported the efficacy of catheter ablation for electrical storms, which was performed within 24h after admission of the patient, and showed a high cumulative mid-term survival (median 15 months) of 91%.

Conclusions

In patients with drug-resistant frequent monomorphic VTs or PVCs triggering VF during an AHF decompensation, an urgent RFCA was completed without any major complications except for a temporary exacerbation of pulmonary congestion, and achieved acute success in 87% of the patients. All targeted PVCs and about 30% of clinically documented monomorphic VTs were eliminated at the sites with Purkinje potentials. Although one-third of the patients required a second session soon after the first, the tachyarrhythmias were completely suppressed in 80% of the patients and became significantly less frequent in the others during hospitalization. The AHF ameliorated after the RFCA in 93% of the study subjects, and 80% were discharged alive. Radiofrequency catheter ablation could be an appropriate therapeutic option for lethal ventricular arrhythmias during an AHF decompensation.

Funding

This work was supported by national grant named MEXT KAKENHI Grant Number 22790735.

Acknowledgements

We thank Mr John Martin for his linguistic assistance.

Conflict of interest: none declared.

References