Stellate ganglion ablation by conventional radiofrequency in patients with electrical storm

Abstract

Aims

We report a series of patients with Electrical Storm (ES) who underwent bilateral stellate ganglion ablation by using conventional radio frequency (RFA).

Methods and results

The procedure was done with fluoroscopic guidance using the COSMAN™ 1A RF Generator and a 22G RF needle (5 cm length and 5 mm active tip). Six patients, four male and two female (mean age 55 ± 7 years and mean LVEF—42 ± 21%) with ES underwent the procedure under fluoroscopic guidance. All patients experienced recurrent ICD shocks or required multiple external defibrillation shocks. There were no procedural complications. All patients survived free of ES at discharge. At a mean follow—up of 22 ± 8months, all were alive free of ES but two patients received appropriate shocks for VT and one patient had VT terminated by ATP.

Conclusion

This small series of cases is a proof of concept that neuromodulation by conventional RFA targeting bilateral stellate ganglia appears safe, feasible, and effective in treating selected unstable patients with ES.

Graphical Abstract

Graphical abstract.

The clinical outcomes of the six patients who underwent SGA (stellate ganglion ablation) are depicted in the figure. All patients were free of ES and VT at the last follow-up. Two patients experienced appropriate shock and one patient had VT terminated by ATP.

What’s new?

• First report of neuromodulation by conventional radiofrequency ablation of stellate ganglion in patients with electrical storm.

• Feasibility, efficacy, and safety of the procedure are shown in a series of six patients.

• This is a cost effective non-surgical stellate ganglion intervention, a feasible alternative to bilateral surgical sympathetic denervation in patients with life-threatening electrical storm.

Various techniques of neuromodulation have been described to treat electrical storm (ES) and refractory VT.^1–7 We describe our initial experience of treating six patients with ES with bilateral stellate ganglion ablation (SGA) using conventional radiofrequency at C7 and T1 levels.

The procedure was done after obtaining informed consent. Ablation was done under fluoroscopic guidance either in single (n = 5) or separate sittings (n = 1) depending on the clinical stability of the patient (Figure 1A and B). The patient was placed in a supine position with his head turned opposite to the procedure site and neck kept in extension. When not ventilated, the neck muscles were relaxed by keeping the patient’s mouth partially open and advising not to speak during the procedure. For targeting C7, anterior paratracheal approach was used to reduce the incidence of injury to the recurrent laryngeal nerve. After infiltrating local anaesthetic agent, the junction between the transverse process and the vertebral body of C7 was profiled by fluoroscopy in Antero-Posterior (AP) view. A 22G RF needle of 5 cm length and 5 mm active tip was inserted and directed towards the transverse process to a depth of ∼3 cm. During the insertion of this needle, the sternocleidomastoid muscle and the carotid artery were retracted laterally to prevent inadvertent puncture. The position of the needle was confirmed by a lateral view. Further correct placement of the needle was confirmed by contrast injection that showed a cephalocaudal dispersion of dye confirming the position of the tip anterior to the longus colli muscle and ruling out intravascular injection. The COSMAN™ 1A RF Generator (Boston Scientific, Marlborough, MA, USA) was used for stimulation and ablation. A sensory (50 Hz, 0.9 V) and a motor (2 Hz, 2 V) stimulation trial were initially performed to rule out phrenic nerve and recurrent laryngeal nerve involvement according to the standard protocol.⁸ In non-ventilated cases, during motor stimulation, the patient was asked to say ‘ee’ and motor function preservation was confirmed. Three lesions were delivered using continuous (conventional) radiofrequency at 50°C, 55°C, and 60°C successively for 90 s each. For the next step, patient was turned to prone position to avoid pneumothorax. T1 vertebra was identified, and the needle was placed at the junction of the vertebral body and the corresponding transverse process. Conventional radiofrequency ablation was performed at 70°C for 60 s each, and three lesions were delivered. The entire procedure was repeated on the right side. The endpoints of ablation were ptosis and miosis (part of Horner’s syndrome).

Figure 1.

(A and B) Fluoroscopic images showing the radiofrequency needle positioning at the junction of the vertebral body with the lateral process at C7 level in AP (left) and lateral views (right).

There were six patients, (four males and two females; mean age—55 ± 7 years and mean Left Ventricular Ejection Fraction (LVEF) —42 ± 21%.) Underlying substrates were ischaemic cardiomyopathy (ICM)—two, non-ischaemic dilated cardiomyopathy (NICM)—two, hypertrophic cardiomyopathy (HCM)—one, and structurally normal heart—one. All patients were initially treated with our institutional protocol of ES management.^9,10 They underwent SGA as they continued to have VT/VF requiring multiple defibrillation shocks.

The procedure was done under general anaesthesia in four (administered as a part of ES management) and under local anaesthesia in two. The mean lesion time was 12 ± 6 min, and the procedure time was 38 ± 9 min. There were no procedure related complications. Horner’s syndrome seen at the end of the procedure was transient and not seen in any of the patients at follow-up.

All patients were rescued from ES and survived to discharge. At a mean follow-up of 22 ± 8 months, two patients received appropriate shocks for VT and in one patient VT was terminated by ATP (Graphical abstract). Two patients underwent electroanatomical mapping and ablation—one had endocardial substrate modification and in the other, triggering ventricular ectopy was targeted (Table 1).

Table 1.

Clinical profile of patients and follow-up

Patient #	1	2	3	4	5	6
Age in years and gender	59 M	60 M	53 M	55 F	63 M	42 F
Clinical presentation	Multiple appropriate ICD shocks	Multiple polymorphic VT/VF episodes	Multiple polymorphic VT/VF episodes	Multiple appropriate ICD shocks	Multiple monomorphic VT episodes	Multiple appropriate ICD shocks for VF
No. of shocks prior to procedure (within 48 h)	14	15	17	18	11	30
VT cycle length (ms)	303	210	205	259	323	203
Substrate	HCM	ICM (post-CABG Day 10)	ICM (post-CABG Day 4)	NICM (sarcoidosis)	NICM	Ectopy-induced torsades
LVEF (%)	66%	25%	18%	55%	28%	60%
AAD used in addition to metoprolol	Lidocaine, disopyramide	Lidocaine, amiodarone, phenytoin	Amiodarone, lidocaine	Amiodarone, lidocaine	Amiodarone, lidocaine	Lidocaine, mexilitine, verapamil
Mechanical ventilation	Yes	Yes	Yes	No	No	Yes
Reason for not performing conventional BCSD	Financial constraints	Family and surgical team not ready for 2nd thoracic procedure	Family and surgical team not ready for 2nd thoracic procedure	To avoid G/A	To avoid G/A	Financial constraints
Follow-up duration (months)	24	28	26	22	25	6
VT/shocks at F/U	One shock at 6 months	None	Three VT episodes terminated by ATP	Three shocks after 7 months	None	None
VT cycle length at F/U (ms)	384	N/A	382	363	N/A	N/A
VT ablation	No	No	No	Yes (at 7 months)	No	Yes (at 24 h)
AAD at last follow-up	None	None	Amiodarone	Amiodarone	None	None

AAD, Anti-arrhythmic drugs; BCSD, Bilateral cardiac sympathetic denervation; F/U, Follow-up; CABG, Coronary Artery Bypass Surgery; G/A, General anaesthesia; HCM, Hypertrophic cardiomyopathy; ICD, Implantable Cardioverter Defibrillator; ICM, Ischaemic cardiomyopathy; LVEF, Left Ventricular Ejection Fraction; NICM, Non-ischaemic cardiomyopathy; VF, Ventricular Fibrillation.

To the best of our knowledge, this is the first report of bilateral SGA using fluoroscopically guided conventional radiofrequency energy in the treatment of ES. We chose neuromodulation in our patients for various reasons. Patient #1 had an apical aneurysm with a thrombus. Patients #2 and #3 were both in perioperative period with haemodynamically unstable VF (Ventricular Fibrillation) and polymorphic VT, respectively, with no definite substrates or targets for ablation. Patient #4 had a substrate (mid-myocardial scar) where ablation results are not very promising and much less in unstable ES patients.¹¹ Patient #5 did not consent to undergo ablation. Patient #6 had frequent episodes of VF, each triggered by a ventricular ectopy and when ES subsided with SGA, the monomorphic ectopy could be ablated.

Our technique has important procedural differences from pulsed radiofrequency ablation that is a percutaneous procedure well established and tested for treating chronic neuropathic pain situations due to spine pathologies.¹² Ablation at C7 level in addition to T1 level ensured a more definitive sympathetic interruption. Restricting the temperatures to <60°C (instead of 70–80°C used for pain management) with a 5 mm RF tipped needle avoided collateral damage to the recurrent laryngeal/phrenic nerves and the brachial plexus. This technique results in only partial denervation, and a permanent damage is difficult to achieve and demonstrate. It has potential advantage over surgical bilateral cardiac sympathetic denervation (BCSD) avoiding general anaesthesia, single lung ventilation, and complications like pneumothorax, haemothorax, and Horner’s syndrome that are reported in 11.5% of the patients.^13,14 Modifications in thoracoscopic surgical techniques have been described that are less invasive but still need general anaesthesia and moreover, stellate ganglia are spared.^15,16

All patients in this series were successfully rescued from ES, and at a mean follow-up of nearly 2 years; there was control of VT events in five out of six patients, and no recurrence of ES. This technique could prove to be more than just a bail-out procedure, though more data are needed to confirm this observation. However, the fact that this procedure had no long-term complications and only transient Horner’s was observed likely indicates that ablation effects at C7 may have been transient.

This small series is a proof of concept that neuromodulation by conventional radiofrequency ablation targeting bilateral stellate ganglia appears safe, feasible, and effective in treating selected unstable patients with ES.

Funding

None declared.

Data availability

Data will be provided on request to the corresponding author by email.