Clinical aspects

Description

There is only one noninvasive vagal nerve stimulation device cleared by the United States Food and Drug Administration for acute and preventive treatment of cluster headache and migraine and is further supported by several randomized control trials and real-world evidence.

Cluster headaches. Two randomized, double-blind, sham-controlled trials (ACT1, ACT2) have reported noninvasive vagal nerve stimulation as safe and efficient in treating acute episodic cluster headaches (27; 14). In ACT1 (NCT01792817), Silberstein and colleagues evaluated 133 subjects (episodic cluster headache n=38, chronic cluster headache n=22, sham n=73) who received noninvasive vagal nerve stimulation (three consecutive 120-second stimulations with 60-second intervals) at the onset of premonitory symptoms or pain. Subjects self-treated up to five cluster headache attacks in the double-blind phase; the primary endpoint was the response rate, defined as the proportion of all subjects who achieved a pain intensity score of 0 or 1 on a 5-point scale (0, no pain; 4, very severe pain) at 15 minutes after treatment initiation for the first cluster headache attack. Treatment response was achieved in 26.7% of noninvasive vagal nerve stimulation-treated subjects and 15.1% of sham-treated subjects (P=0.1). Response rates were significantly higher with noninvasive vagal nerve stimulation than with sham for the episodic cluster headache cohort (34.2% vs. 10.6%; P=0.008) but not for the chronic cluster headache cohort (13.6% vs. 23.1%; P=0.48) (27). In the ACT2 study (NCT01958125), participants were randomly assigned to the noninvasive vagal nerve stimulation group (episodic cluster headache=14, chronic cluster headache=34) and the sham group (episodic cluster headache=13, chronic cluster headache=31) for a 2-week double-blinded period. The primary endpoint was the proportion of all treated cluster headache attacks that achieved pain-free status within 15 minutes of treatment initiation. The study demonstrated no significant difference in primary endpoint achievement between noninvasive vagal nerve stimulation (14%) and sham (12%) treatments. Similarly, no significant differences were reported between noninvasive vagal nerve stimulation (5%) and sham (13%) in the chronic cluster headache subgroup. However, noninvasive vagal nerve stimulation (48%) was superior to sham (6%; p< 0.01) in the episodic cluster headache subgroup (14).

The efficacy of noninvasive vagal nerve stimulation for chronic cluster headache prevention was explored in an open-label prospective randomized clinical trial (PREVA) by Gaul and colleagues comparing standard of care (n=49) and adjunctive noninvasive vagal nerve stimulation (n=48) treatments (12). After a 4-week randomized phase, the groups were analyzed for achieving the primary endpoint (a reduction in the mean number of cluster headache attacks per week) along with response rate, abortive medication use, and safety. In the randomized phase, the standard of care plus noninvasive vagal nerve stimulation group had a significantly greater reduction in cluster headache attacks per week (-5.9 vs. -2.1) and a mean of 3.9 fewer attacks per week (p=0.02). The adjunctive treatment group also demonstrated a higher 50% responder rate (40.0% vs. 8.3%; p< 0.001) (12). After the 4-week randomization phase, 92 of the study participants entered a 4-week extension phase receiving both standard of care and noninvasive vagal nerve stimulation interventions. In a post hoc analysis, the response rate was also significantly reduced at the end of the randomized period in the modified intent-to-treat (mITT) population. Compared adjunctive nVNS to standard of care alone, 50% response rates were 49% vs. 9% (P <  0.001), and 75% response rates were 22% vs. 2% (P = 0.009) (13). In a retrospective analysis conducted in the United Kingdom on 30 patients with cluster headache, after an evaluation period of 3 to 6 months, there was a significant decrease in mean attack frequency from 26.6 attacks per week to 9.5 attacks per week (p< 0.01). Mean attack duration also decreased from 51.9 minutes to 29.4 minutes (p< 0.01), and mean attack severity decreased from 7.8 to 6.0 (p< 0.01) on a 10-point scale (19). An open-label study of nVNS in patients with medically refractory CCH shows greater than 50% reduction in headache frequency in 42.5% patients (29). These results are in alignment with the metaanalysis, which showed responder proportion of 0.35 (95% CI 0.07 to 0.69, n = 137) and a mean reduction in headache frequency of 35.3 attacks per month (95% CI 11.0 to 59.6, n = 108), from a baseline of 105 (±22.7) attacks per month. The study suggests the potential benefit of nVNS in CCH, but randomized sham-controlled trials are needed to confirm the beneficial response of VNS reported in some, but not all open-label studies (29).

Table 1. Summary of Noninvasive Vagal Nerve Stimulation Randomized Controlled Trials on Cluster Headache

Migraine. In 2014, Goadsby and colleagues conducted an open-label pilot study to assess the efficacy of noninvasive vagal nerve stimulation for the acute treatment of migraine. The study assessed the response of noninvasive vagal nerve stimulation monotherapy on 19 patients with moderate or severe headaches. They reported 21% (n=4) pain-free response and 47% (n=9) with pain relief after two hours of noninvasive vagal nerve stimulation therapy (15). This pilot study was followed up in 2018 by PRESTO, a large randomized, double-blinded, sham-controlled clinical trial (NCT02686034). In a total of 248 episodic migraine patients, 12.7% of the noninvasive vagal nerve stimulation (n=120) group had pain freedom in comparison to the 4.2% response of the sham (n=123) group at 30 minutes (p=0.012); similar response was seen at 60 minutes (21.0% vs 10.0%; p=0.023) but not at 120 minutes (30.4% vs. 19.7%; p=0.067). Noninvasive vagal nerve stimulation also exhibited significant pain relief at 120 minutes (40.8% vs. 27.6%, p=0.03) while being safe and well-tolerated. Through this study, Tassorelli and colleagues provided evidence of the efficacy and safety of noninvasive vagal nerve stimulation devices for acute episodic migraines (31).

For migraine prevention, three randomized controlled trials used transcervical vagus nerve stimulation, and two used transauricular vagus nerve stimulation. In the EVENT trial, the efficacy and benefits of noninvasive vagal nerve stimulation on chronic migraine were tested in an early prospective, double-blinded, sham-controlled pilot study (NCT01667250). Fifty-nine patients with a mean headache frequency of 21.5 days per month were randomized into noninvasive vagal nerve stimulation (n=30) and sham (n=29) treatment groups. After two months, the mean number of headache days changed by -1.4 for noninvasive vagal nerve stimulation and -0.2 for the sham group (p=0.56). Of the 30 participants enrolled in the noninvasive vagal nerve stimulation group, 15 completed a 6-month open-label treatment period. In those who completed a 6-month open-label extension (n=15), the mean number of headache days decreased significantly (-7.9) from baseline (95% CI -11.9 to -3.8; p< 0.01) (26). In the PREMIUM trial, the efficacy of noninvasive vagal nerve stimulation on episodic migraine was evaluated by a randomized, double-blinded, sham-controlled study (10). After a 12- week randomized treatment of the noninvasive vagal nerve stimulation group (n=165) and the sham group (n=167), no significant mean reductions in migraine days per month (primary endpoint) were reported. A reduction of 2.26 migraine days per month was observed in the noninvasive vagal nerve stimulation group (baseline, 7.9 days) and 1.80 days for the sham group (baseline, 8.1 days) (p=0.15). However, in the mITT population (patients who were 67% or greater adherent to stimulation), patients showed a significantly greater reduction in the monthly migraine days (-2.27 vs. -1.53, p=0.043), monthly headache days (-2.85 vs. -1.99, p=0.045), and acute medication use days (-1.94 vs. -1.14, p=0.039). Interestingly, based on this robust response seen in the sham group in the PREMIUM study, it was theorized that the sham device might have stimulated the vagus nerve. Schroeder and colleagues demonstrated that both noninvasive vagal nerve stimulation and sham (0.1hz) produced stimulation, but no stimulation significantly reduced ipsilateral trigeminal autonomic reflex induced by kinetic oscillation stimulation (25).

Several limitations seen in the original PREMIUM study were addressed in a second large, 12-week, randomized double-blinded sham-controlled trial (PREMIUM II, NCT03716505). PREMIUM II applied ipsilateral stimulations to the side of predominant pain compared to the original study that used bilateral stimulations. The study also expanded to both episodic and chronic migraine subjects (8 to 20 monthly headache days) and utilized a modified inactive sham device to avoid sham stimulation. Of 231 randomized patients, 113 (active=56, sham=57; chronic migraine=23, episodic migraine=33) completed 70 or more days of the 12-week double-blind period and adhered 66% or more to treatment each week. The study demonstrated a mean reduction in monthly migraine days of 3.12 days in the active group and 2.29 in the sham group (p=0.2329). They also reported a greater 50% response rate in the active group than the sham group (44.87% vs. 26.81%, p=0.0481). The trial was terminated early due to the COVID-19 pandemic, leading to a smaller statistical target population. The study reinforced the efficacy and safety of noninvasive vagal nerve stimulation devices for migraine headaches (22).

In addition to transcervical vagus nerve stimulation, there are two randomized controlled trials evaluated transauricular vagus nerve stimulation for migraine prevention. Straube and colleagues investigated the therapeutic effects of NEMOS® on patients with chronic migraine by applying the device at the left tragus and stimulating the auricular branch of the vagus nerve. The device was applied to 46 patients over three months, for 4 hours per day, delivering 25 Hz to 1 Hz stimulation. The study showed a significant reduction in the number of headache days in the active control group that received 1 Hz stimulation as compared to the 25 Hz group (-7.0 ± 4.6 vs. -3.3± 5.4 days, p=0.035). They also assessed headache-related disability by the Headache Impact Test (HIT-6) and the Migraine Disability Assessment (MIDAS) questionnaires, both showing significant improvement (30). Another randomized study was conducted by Zhang and colleagues where 70 patients with episodic migraine received four weeks of treatment with taVNS (30 minutes, total 12 sessions) or sham. The taVNS group significantly reduced the number of migraine days (-2.5 vs. -0.7, p=0.024), pain intensity (-17.4 vs. -4.1, p=0.008), and attack times (-1.5 vs. 0.4, p=0.015) as compared to the sham taVNS. Their study further analyzed the modulatory effect of taVNS devices on the thalamocortical circuits, revealing taVNS induced increased connectivity between the anterior cingulate cortex and the motor-related thalamus subregion (37).

Table 2. Summary of Noninvasive Vagal Nerve Stimulation Randomized Controlled Trials on Migraine

Other headache disorders. gammaCore received United States Food and Drug Administration clearance for patients with hemicrania continua based on a few case reports. Eren and colleagues reported a man suffering from left-sided hemicrania continua who developed contraindication to indomethacin (11). The patient received noninvasive vagal nerve stimulation using the PREVA protocol and reported an immediate decline of pain intensity from 8/10 to 5/10 during exacerbation periods. He also reported improved quality of life and subjective reduction in the background pain intensity. Similarly, Tso and colleagues described 10 patients with hemicrania continua and paroxysmal hemicrania who could not tolerate indomethacin and used noninvasive vagal nerve stimulation as an adjunctive therapy (32). In patients with hemicrania continua, 78% reported reduced severity of continuous pain. In addition, 67% of patients reported benefits such as being attack-free, reduced attack frequency, reduced severity, and shorter duration of attacks. Patients with primary cough headaches demonstrated improvement with noninvasive vagal nerve stimulation. Moreno-Ajona and colleagues reported a man with primary cough headaches being treated with noninvasive vagal nerve stimulation after developing contraindications to indomethacin (20). There was a dramatic improvement over six weeks; he reported complete relief afterward. The treatment efficacy of noninvasive vagal nerve stimulation devices for vestibular migraine was also demonstrated in a case series reported by Beh (03). The patients received bilateral 120-second stimulations on both sides of the neck and were asked to grade their vertigo and headache using a 10-point visual analog scale. The patients reported a difference in their average vertigo severity from 5 to 1.5 and mean headache severity from 4 to 0.7 after noninvasive vagal nerve stimulator use. With these findings, gammaCore was cleared by the US FDA for hemicrania continua and paroxysmal hemicrania but not yet for vestibular migraine and primary cough headaches.

Indications

gammaCore Sapphire™ (electroCore, Inc., Rockaway, NJ) is a handheld trans-cervical noninvasive vagal nerve stimulation neuromodulation device that the US FDA has cleared for cluster headaches for both acute and preventive treatment. Indications are as follows:

tVNS® L (tVNS Technologies GmbH, Bayern, Germany), previously knowns as NEMOS, has also received European CE marking for multiple conditions. On the website shop.t-VNS.com, the following indications are listed without any link for references: anxiety, asthma, atrial fibrillation, autism, cognitive impairment, Crohn disease, depression, epilepsy, fibromyalgia, inflammation, migraines, Parkinson disease, Prader-Willi syndrome, sleep disorders, stroke, and tinnitus.

Contraindications

Generally, noninvasive vagal nerve stimulation devices are very safe for use. However, noninvasive vagal nerve stimulation is not applicable in patients with altered anatomy (eg, cervical vagotomy, deformed ear, local skin infection). Similar to invasive vagus nerve stimulation, patients with pre-existing swallowing, cardiac, or respiratory difficulties may experience worsening symptoms. nVNS should be removed from the skin or turned off immediately if any of these symptoms occur.

On gammaCore safety information, contraindication includes:

Adverse effects

Noninvasive vagal nerve stimulation is a safe and well-tolerated treatment option, with some mild side effects that have been reported over the years (24). Redgrave and colleagues conducted a systematic review of all the studies that deployed noninvasive vagal nerve stimulation in humans. They reported local skin irritation as the most common side effect, seen in 18.2% of participants. Local skin reactions included tingling or pain around stimulation sites, with only a few participants reporting other skin irritations, such as itchiness or redness. Skin irritation was followed by headaches, seen in 3.6% of the participants. Other rare side effects were also reported, such as heart palpitations, nasopharyngitis, and facial drooping. Gastrointestinal issues such as nausea or vomiting were reported.

Special considerations

The safety and efficacy of gammaCore have not been evaluated in the following patients:

Scientific basis

Researchers have proposed multiple hypotheses regarding the neural activation effects of noninvasive vagal nerve stimulation. Silberstein and colleagues contemplated four core areas whose overlap and interplay provide mechanistic explanations noninvasive vagal nerve stimulation effectiveness. These four core areas include autonomic nervous system functions, cortical spreading depression, nociceptive modulation, and neurotransmitter regulation (28). The autonomic nerve and its direct and indirect connections with the vagus nerve in the brainstem form the basis of pain pathways in the head; these pain pathways are modulated by noninvasive vagal nerve stimulation, resulting in inhibition of trigeminocervical neuron firing. fMRI findings from Zhang and colleagues demonstrated increased connectivity between thalamic motor speed with the rostral anterior cingulate cortex in the rostral anterior cingulate cortex group after treatment (37). They also reported decreased connectivity between the thalamic subregions and brain regions associated with pain sensitivity, such as the postcentral gyrus, in comparison to the sham group. The initiation and propagation of cortical spreading depression are affected by noninvasive vagal nerve stimulation as demonstrated by rat models, which had a greater than two-fold increase in the electrical stimulation needed to induce a cortical spreading depression after the use of a noninvasive vagal nerve stimulation device (08). Furthermore, in rat models of head pains, including migraine-like and cluster-like pains, vagus nerve stimulation has suppressed the acute nociceptive activation of trigeminocervical neurons (01). The nociceptive modulation of noninvasive vagal nerve stimulation devices was shown in two rodent models of episodic migraines as well (09). Kinfe and colleagues also proposed that stimulating the vagus nerve opposes the decline of thalamocortical activity seen in patients with migraine and sleep disorders (17). Noninvasive vagal nerve stimulation may increase neural pathway activity in the locus coeruleus, enhancing memory performance. This was supported by Burger and colleagues’ suggestion of the role of noninvasive vagal nerve stimulation in the activation of the limbic areas of the brain, including the amygdala and hippocampus (05).

Vagal nerve stimulation has shown potential applications in other fields outside of headache in animal models. Using optogenetics, vagal nerve stimulation has been shown to drive precise temporal modulation of neurons in the primary motor cortex (M1), which responds to behavioral outcomes, suggesting that vagal nerve stimulation may accelerate motor refinement in M1 via cholinergic signaling (04). tcVNS device affects brain activity and heart rate variability (HRV) in dogs, which can be detected by EEG and Holter monitoring. tcVNS was also shown to result in a significant decrease in the alpha (p < 0.01), theta (p = 0.01), and beta (p = 0.035) frequencies poststimulation, as well as an increase in HRV measured by the standard deviation of the inter-beat (SDNN) index (p < 0.01) and a decrease in mean heart rate (p < 0.01) (07). Despite being a pilot study, the study implements the future possibilities of using EEG and Holter monitoring as means of monitoring the effect of vagal nerve stimulation devices. Vagal nerve stimulation has also been shown to promote the resolution of inflammation in mice. Vagal nerve stimulation increases local levels of specialized pro-resolving mediators and accelerates the resolution of inflammation in zymosan-induced peritonitis by a mechanism that involves Alox15 and requires the cholinergic α7-nicotinic acetylcholine receptor subunit (α7nAChR) (06).

Vagal nerve stimulation may have a modulatory and potential therapeutic role in stomach-brain coupling and autoimmune diseases, respectively. An ongoing clinical trial SLE-VNS study is evaluating the effect of tVNS in measures of fatigue, clinical disease activity, pain, and organ function while it is also hypothesized to reduce inflammation and hence improve quality of life as a part of adjuvant therapy for patients with SLE (38). The vagal signaling between digestive organs and the brain plays a critical role in maintaining energy homeostasis. In a single-blinded randomized crossover study, taVNS increased stomach-brain coupling in the nucleus of the solitary tract (NTS) and midbrain regions, as well as across the brain. Notably, taVNS-induced changes in cortical coupling were predominantly observed in transmodal regions, which are associated with higher-order sensory processing and integration. These changes in cortical coupling were accompanied by changes in hunger ratings, indicating that taVNS may modulate the subjective metabolic state. By increasing stomach-brain coupling via an NTS-midbrain pathway that signals gut-induced reward, taVNS may help to better understand the role of vagal afferents in orchestrating the recruitment of the gastric network (21).