Effect of Transcutaneous Magnetic Stimulation in Patients With Ventricular Tachycardia Storm: A Randomized Clinical Trial

Key Points

Question  What is the effect of transcutaneous magnetic stimulation (TcMS) of the left stellate ganglion in patients with ventricular tachycardia (VT) storm?

Findings  This double-blind, sham-controlled randomized clinical trial of TcMS included 26 patients with VT storm. Treatment with a single session of TcMS vs sham stimulation resulted in recurrent VT in 29% of patients in the TcMS group vs 58% in the sham group; this difference was not statistically significant, and treatment with TcMS resulted in fewer episodes of VT during the 72 hours following randomization.

Meaning  These findings suggest that treatment with TcMS is safe and may be an effective option to stabilize patients in VT storm.

Importance  Autonomic neuromodulation provides therapeutic benefit in ventricular tachycardia (VT) storm. Transcutaneous magnetic stimulation (TcMS) can noninvasively and nondestructively modulate a patient’s nervous system activity and may reduce VT burden in patients with VT storm.

Objective  To evaluate the safety and efficacy of TcMS of the left stellate ganglion for patients with VT storm.

Design, Setting, and Participants  This double-blind, sham-controlled randomized clinical trial took place at a single tertiary referral center between August 2019 and July 2021. The study included 26 adult patients with 3 or more episodes of VT in 24 hours.

Interventions  Patients were randomly assigned to receive a single session of either TcMS that targeted the left stellate ganglion (n = 14) or sham stimulation (n = 12).

Main Outcomes and Measures  The primary outcome was freedom from VT in the 24-hour period following randomization. Key secondary outcomes included safety of TcMS on cardiac implantable electronic devices, as well as burden of VT in the 72-hour period following randomization.

Results  Among 26 patients (mean [SD] age, 64 [13] years; 20 [77%] male), a mean (SD) of 12.7 (10.3) episodes of VT occurred within the 24 hours preceding randomization. Patients had recurrent VT despite taking a mean (SD) of 2.0 (0.6) antiarrhythmic drugs (AADs), and 11 patients (42%) required mechanical hemodynamic support at the time of randomization. In the 24-hour period after randomization, VT recurred in 4 of 14 patients (29% [SD 47%]) in the TcMS group vs 7 of 12 patients (58% [SD 51%]) in the sham group (P = .20). In the 72-hour period after randomization, patients in the TcMS group had a mean (SD) of 4.5 (7.2) episodes of VT vs 10.7 (13.8) in the sham group (incidence rate ratio, 0.42; P < .001). Patients in the TcMS group were taking fewer AADs 24 hours after randomization compared with baseline (mean [SD], 0.9 [0.8] vs 1.8 [0.4]; P = .001), whereas there was no difference in the number of AADs taken for the sham group (mean [SD], 2.3 [0.8] vs 1.9 [0.5]; P = .20). None of the 7 patients in the TcMS group with a cardiac implantable electronic device had clinically significant effects on device function.

Conclusions and Relevance  In this randomized clinical trial, findings support the potential for TcMS to safely reduce the burden of VT in the setting of VT storm in patients with and without cardiac implantable electronic devices and inform the design of future trials to further investigate this novel treatment approach.

Trial Registration  ClinicalTrials.gov Identifier: NCT04043312

Ventricular tachycardia (VT) storm is associated with considerable morbidity and mortality.¹ Autonomic neuromodulation via local blockade of the left stellate ganglion provides therapeutic benefit by reducing cardiac sympathetic input in patients with VT storm.^2-5 This strategy is invasive and limited by associated risks and the need for technically skilled clinicians.

Transcutaneous magnetic stimulation (TcMS) can noninvasively and nondestructively modulate nervous system activity in a variety of disease states.^6-8 Magnetic stimulation has been shown to modify arrhythmia risk by targeting cardiac sympathetic innervation in animal models.^9,10 We had previously performed a feasibility study of TcMS in patients with VT storm, which demonstrated a lower burden of VT with no adverse events.¹¹ This double-blind, randomized pilot study was conducted to characterize the effects of a single session of TcMS for patients with VT storm.

This single-center randomized clinical trial was conducted at the Hospital of the University of Pennsylvania. The institution’s institutional review board approved this study and its protocol (Supplement 1), and all patients or their surrogate decision makers provided informed consent. Consent was written when possible and verbal when written consent was not possible. Consolidated Standards of Reporting Trials (CONSORT) reporting guidelines were followed.

Eligible candidates were 18 years or older with at least 3 episodes of sustained VT in the preceding 24 hours. Sustained VT was defined as lasting 30 seconds or longer or resulting in antitachycardia therapy (ie, shock to restore sinus rhythm or antitachycardia pacing). Exclusion criteria included the presence of a durable implanted ventricular assist device, metal in the head or neck (except the mouth), implanted medication pump or brain stimulator, cochlear or ocular implant, pregnancy, or history of cancer in the region of stimulation. Patients were not enrolled if there was a plan to perform a VT ablation within 24 hours, given the potential for confounding the primary end point.

Patients were randomized in a 1:1 ratio to receive a single session of TcMS or sham stimulation. The patient, treatment team, and physician assessing arrhythmic outcomes were blinded to the treatment arm. All patients with a cardiac implantable electronic device (CIED) underwent device interrogation and reprogramming using protocols previously used for magnetic resonance imaging examinations.¹² Baseline settings were restored, and additional device reprogramming was not performed during the 72 hours following randomization.

A figure-of-8 TcMS coil attached to a Magstim Super Rapid magnetic stimulation system (Magstim Company Limited) was positioned in approximation of the left stellate ganglion. For patients randomized to receive TcMS, stimulation was delivered at 80% of the local motor threshold and frequency of 0.9 Hz for 60 minutes. For patients randomized to receive sham stimulation, sham therapy was delivered for 60 minutes. Blood pressure, heart rate, and bilateral index finger perfusion index via pulse oximeter (Innovo) were assessed during stimulation. A 12-lead electrocardiogram and bilateral palmar skin temperature via infrared thermometer (AmpMed) were assessed before and after completion of the protocol. For nonsedated patients, the presence or degree of discomfort was assessed.

The primary outcome was freedom from sustained VT in the 24-hour period following randomization. Secondary outcomes included increase in pulse oximeter and palmar skin temperature, change in CIED parameters, and the burden of sustained VT in the 72 hours following randomization.

The sample-size calculations assumed that 80% of patients in the sham stimulation group and 20% of patients in the TcMS group would have recurrent VT during the 24-hour period after randomization. Nonparametric statistical tests were used when the data were not normally distributed. Patients were analyzed according to their allocated treatment group. Poisson regression was used to evaluate the burden of recurrent VT. A 2-sided P value  < .05 was considered statistically significant. Owing to multiple comparisons resulting in the potential for inflated type I error, secondary analyses should be considered exploratory. Analyses were performed using Stata, version 12 (StataCorp).

Between August 2019 and July 2021, a total of 30 patients were eligible and 4 patients declined participation (Figure 1). The remaining 26 patients (mean [SD] age, 64 [13] years; 20 [77%] male) were randomized to receive TcMS (n = 14) or sham stimulation (n = 12). During the 24-hour period prior to randomization, patients had a mean (SD) of 12.7 (10.3) episodes of sustained VT, which had recurred despite taking 2.0 (0.6) antiarrhythmic drugs (AADs) (Table). A patient in each group underwent percutaneous stellate ganglion block with lidocaine 1% within 72 hours after randomization.

During the 24-hour period following randomization, sustained VT recurred in 4 of 14 patients (29% [SD 47%]) in the TcMS group and in 7 of 12 patients (58% [SD 51%]) in the sham stimulation group (Figure 2A). The difference was not statistically significant (P = .20).

During the 72-hour period following randomization, the mean (SD) number of episodes of sustained VT was 4.5 (7.2) in patients in the TcMS group compared with 10.7 (13.8) in patients in the sham stimulation group (Figure 2B). Treatment with TcMS was associated with fewer episodes of VT compared with sham stimulation during this period with an incidence rate ratio (IRR) of 0.42 (P < .001). Patients in the TcMS group had fewer episodes of VT in the first 24 hours after randomization (mean [SD], 1.1 [1.9] vs 6.0 [9.6] episodes; IRR, 0.18; P < .001) but not in the 25- to 72-hour period (IRR, 0.88; P = .60). Patients in the TcMS group were taking fewer AADs 24 hours after randomization compared with baseline (mean [SD], 0.9 [0.8] vs 1.8 [0.4]; P = .001) whereas there was no difference in the number of AADs taken for the sham group (mean [SD], 2.3 [0.8] vs 1.9 [0.5]; P = .20) (eFigure in Supplement 2).

There were no acute safety concerns related to TcMS therapy. No patient reported pain during the procedure.

There were CIEDs present in 7 of 14 patients in the TcMS group. No clinically significant adverse events occurred during stimulation. No device parameters, including sensing, pacing thresholds, and lead impedances, had clinically significant changes.¹² Magnet mode was not activated for any patient, and TcMS was not detected by the device.

Among the 11 patients (8 [57%] in the TcMS group and 3 [25%] in the sham stimulation group) not receiving treatment with intravenous vasopressors, 4 patients in the TcMS group and no patients in the sham stimulation group had evidence of peripheral blockade. None of these 4 patients had sustained or nonsustained VT within 24 hours after randomization.

This prospective randomized, sham-controlled trial evaluated the safety and efficacy of TcMS among patients with VT storm. There was no statistically significant difference in 24-hour freedom from VT after randomization. There were no considerable safety concerns with TcMS, and there was no evidence of interference with CIEDs. The present study population reflects the high-risk nature of VT storm with a rate of in-hospital mortality greater than 30%. Prior to randomization, patients had recurrent VT despite treatment with multiple AADs and deep sedation with mechanical hemodynamic support in nearly half of the patients.

Although 24-hour freedom from VT was not different between groups, there was a substantial reduction in the number of episodes of VT in the 72-hour period following randomization in patients who received TcMS that was driven by a reduction in the initial 24-hour period postrandomization. The subset of patients with evidence of unilateral peripheral vasodilation consistent with sympathetic blockade remained free of sustained or nonsustained VT for greater than 24 hours after randomization. Nevertheless, the observation that not all patients treated with TcMS demonstrated evidence of sympathetic blockade reflects the need for improved stimulation techniques and assays for successful modulation.

Interruption of sympathetic cardiac innervation to treat ventricular arrhythmias has been explored using invasive methods.^5,13 Currently available approaches are limited by associated procedural risks and the need for technically skilled clinicians.⁴ Transcutaneous magnetic stimulation is a well-developed technique that exploits the plasticity of a patient’s nervous system and can promote excitation or inhibition depending on stimulation parameters.^14,15 This treatment is valuable not only because of its ability to modify neural circuitry, but also because it is uniquely noninvasive and nondestructive.

Although attempts at blinding were made, TcMS resulted in pacing spikes on telemetry that could have been identified. In theory, this may have influenced management decisions and led to incomplete blinding.

This study was underpowered to detect a difference in the primary end point of 24-hour freedom from VT owing to a lower-than-expected rate of VT in the control group. Additionally, owing to the small sample size, there were important differences between groups, including the baseline use of deep sedation, vasopressors, and mechanical support. The effect of these differences on the study outcomes is unclear, but this emphasizes the need for a larger multicenter study. Future studies must additionally evaluate optimal stimulation parameters, including dose, duration, and potential repeated stimulation procedures in addition to comparing TcMS with percutaneous stellate ganglion blockade.

Although no adverse events occurred, conclusions regarding the safety of TcMS in patients with CIEDs are limited owing to the small number of patients with devices in the TcMS group. In addition, a small number of patients had physiologic data available for analysis, and interpretation based on these findings is limited.

In this randomized clinical trial of patients with VT storm, results demonstrated that TcMS has the potential to safely reduce the burden of VT. These findings should inform future investigations of the optimal strategies for TcMS.

Accepted for Publication: December 20, 2021.

Published Online: February 16, 2022. doi:10.1001/jamacardio.2021.6000

Corresponding Author: Timothy M. Markman, MD, Division of Cardiology, University of Pennsylvania, Hospital of the University of Pennsylvania, 3400 Civic Center Blvd, Philadelphia, PA 19146 (timothy.markman@pennmedicine.upenn.edu).

Author Contributions: Drs Markman and Nazarian had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis. Drs Hamilton and Nazarian contributed equally to this work.

Concept and design: Markman, Hamilton, Nazarian.

Acquisition, analysis, or interpretation of data: All authors.

Drafting of the manuscript: Markman.

Critical revision of the manuscript for important intellectual content: All authors.

Statistical analysis: Markman, Linn, Nazarian.

Obtained funding: Marchlinski, Nazarian.

Administrative, technical, or material support: Pothineni, Zghaib, Amankwah, Kumareswaran, Lin, Hamilton, Nazarian.

Supervision: Marchlinski, Hamilton, Nazarian.

Conflict of Interest Disclosures: Dr Santangeli reported grants from Biosense Webster and personal fees from Abbott, Medtronic, and Baylis Medical outside of the submitted work. Dr Hamilton reported grants from the National Institutes of Health, and personal fees from Starfish Neuroscience and Alexion Pharmaceuticals outside of the submitted work. Dr Nazarian reported grants from Biosense Webster, Imricor, Siemens, and ADAS software as well as personal fees from Circle Cardiovascular Imaging and CardioSolv outside of the submitted work. No other disclosures were reported.

Data Sharing Statement: See Supplement 3.