Vagus nerve stimulation (VNS): recent advances and future directions.

Over the past 50 years, the number of people who are overweight or obese has risen dramatically in most parts of the world [ 69 ]. Obesity often carries with it other comorbidities and increases the risk of other metabolic and cardiovascular diseases. The increased mortality and overall cost burden to global healthcare systems brings about the need for solid and viable treatments for obesity. iVNS has been studied as a potential therapeutic option for patients suffering from obesity, and the evidence is discussed in this section. Several methods have been used to investigate of iVNS for obesity. Along with implanted cervical and noninvasive cutaneous, another form of iVNS, known as infradiaphragmatic iVNS or vagal nerve blocking, has also been explored as a potential treatment option. Vagal nerve blocking requires laparoscopic surgery in which electrodes are placed around the anterior and posterior vagus nerves at the gastroesophageal junction [ 70 ]. These electrodes are connected to a pulse generator that is implanted subcutaneously on the thoracic side wall. In a study originally designed to elucidate the efficacy of invasive cervical iVNS for depression (parameters similar to other depression studies: duty cycle of 30 s on, 5 min off; 250–500 µs pulse width; 30 Hz frequency; 0.25–1.5 mA current amplitude), depressed patients lost an average of 7 kg, with 2–3-point reductions in body mass index (BMI) [ 71 ]. The Maestro System is a vagal nerve blocking device that has been FDA-approved for the treatment of obesity. The system uses a duty cycle of 5 min on, 5 min off, and treats within a range of 1–6 mA at a frequency of 5000 Hz [ 53 ]. Four case series were published between 2008 and 2017, all with positive findings denoted by percent excess weight loss (%EWL). In these case series, the %EWL ranged between 14.2% and 22% [ 72 , 73 , 74 , 75 ]. The success in the early case series led to the completion of three RCTs from 2012 to 2016 [ 70 , 76 , 77 ]. Most of these studies included a behavioral intervention, such as weight management education or counseling, adjunctive to active or sham iVNS treatment. The first RCT, led by Sarr and colleagues, found no significant difference in %EWL between active and sham stimulation, but demonstrated a significant reduction in EWL over time [ 76 ]. The other two RCTs found significant differences in mean %EWL between active and sham conditions, with %EWL ranging from 24.4% to 33% in the active groups and 15.9% to 19% in the sham groups. In one study in particular, response rates were as high as 59% in the active group and 41% in the sham groups [ 77 ]. It is important to note the side effect profile of the abdominal vagus system, which is associated with higher rates of gastrointestinal side effects (heartburn, nausea, dysphagia, and abdominal pain) than other methods of iVNS. This is likely due to the location of implantation and stimulation. Based on the evidence discussed in this section, iVNS was approved for the treatment of morbid obesity by the FDA in 2015. A separate study explored the potential of noninvasive VNS as a treatment for glucose intolerance, collecting measures of BMI as secondary outcomes [ 78 ]. The study found no significant differences in BMI between active and sham stimulation. However, they found that 12 weeks of postprandial taVNS significantly reduced 2-h glucose tolerance and systolic blood pressure. While newer, noninvasive VNS modalities offer hope that clinicians will one day be able to offer these simpler, safer options to patients, additional data are needed to understand whether they are as clinically effective in the treatment of obesity as their invasive counterparts.

The earliest attempt to administer electricity to the vagus nerve came from New York in the late 1800s (see Fig.  1 ). A neurologist named James Leonard Corning invented a fork-like device to treat epilepsy based on the contemporary idea that seizures were due to abnormal cerebral blood flow [ 14 ]. The “Corning fork” functioned via two mechanisms: (1) mechanical compression and (2) electrical stimulation of the carotid sheath. Although the method fell out of favor, interest in electrical stimulation of the vagus nerve was revived in the twentieth century. A series of trials in the early twentieth century began to elucidate the neural effects of electrically stimulating the vagus nerve [ 15 , 16 ]. Following Bailey and Bremer’s seminal trial that demonstrated the ability of VNS to synchronize activity in the feline cortex [ 15 ], Dell and Olson demonstrated the localized effects of VNS in the feline brain [ 16 ]. Building on prior experiments that started decades earlier, MacLean demonstrated the ability of VNS to modulate activity in the cingulate cortex in primates in 1980 [ 17 ]. Fig. 1 History of VNS timeline showing the early VNS experiments in animal models, leading to the first experiment in humans. Over the past few decades, VNS has gained FDA approval for multiple diseases, including medication-refractory epilepsy, treatment-resistant depression, obesity, cluster headache, and post-stroke motor rehabilitation. As noted in the figure, all of the FDA approvals are for implanted VNS with the exception of the approval for cluster headache (which is for a form of noninvasive VNS called transcutaneous cervical vagus nerve stimulation, or tcVNS) History of VNS timeline showing the early VNS experiments in animal models, leading to the first experiment in humans. Over the past few decades, VNS has gained FDA approval for multiple diseases, including medication-refractory epilepsy, treatment-resistant depression, obesity, cluster headache, and post-stroke motor rehabilitation. As noted in the figure, all of the FDA approvals are for implanted VNS with the exception of the approval for cluster headache (which is for a form of noninvasive VNS called transcutaneous cervical vagus nerve stimulation, or tcVNS) The modern invasive VNS device, developed by Zabara and colleagues in the 1980s, has three main components: (1) a pulse generator that is implanted subcutaneously in the patient’s chest wall, (2) an electrode cuff wrapped around the cervical bundle of the left vagus nerve, and (3) a wire connecting the pulse generator and electrode cuff [ 18 ]. A surgeon performs the implantation in an ambulatory setting and, after 2 weeks, activates the device. Over the following weeks and months, a specialized provider (usually a neurologist or psychiatrist) programs the device, adjusting parameters to treat the patient’s symptoms and minimize side effects [ 19 ]. Since the 1980s, the science supporting the therapeutic potential of implanted VNS (iVNS) has been steadily growing for a variety of medical and neuropsychiatric illnesses, including epilepsy, depression, obesity, headaches, other pain-related disorders, inflammatory disorders, and cardiovascular disease. Most of our current understanding comes from animal models, while further study of these indications in humans has been limited by the expensive and invasive nature of the implanted cervical device. In recent years, novel noninvasive VNS methods have been developed, offering the opportunity to translate work from animal models to humans [ 23 ]. To date, two forms of noninvasive VNS have being used for clinical and research purposes. One of the devices involves the noninvasive stimulation of the vagus nerve through the neck, a technique known as transcutaneous cervical vagus nerve stimulation (tcVNS). The other device stimulates the auricular branch of the vagus nerve (ABVN) via electrodes that are placed on the ear, known as transcutaneous auricular vagus nerve stimulation (taVNS). Although believed to mimic the effects of iVNS, it is still unclear whether iVNS and taVNS or tcVNS produce identical effects. When applicable, this review will specify which of the forms of VNS (iVNS, taVNS, or tcVNS) are used in the sections below.

VNS for post-stroke motor rehabilitation is a brilliant example of translating VNS paradigms from animal models to humans. Stroke is the second leading cause of death and disability worldwide, and prevalence is expected to grow by 2030 [ 79 ]. Motor rehabilitation is effective in helping stroke patients regain function and improving quality of life. However, many patients live with deficits beyond the rehabilitation period. Novel methods to enhance the effects of physical rehabilitation in the post-stroke period are essential in reducing this burden on stroke survivors, caregivers, and society. Based on promising preclinical evidence, invasive and noninvasive forms of VNS have now been studied as potential treatments for post-stroke recovery in humans. Invasive cervical iVNS trials evaluated the impact of iVNS paired with physical rehabilitation versus physical rehabilitation with sham stimulation (Table  3 ). Stimulation parameters were consistent across studies; over 6 weeks, patients received 18 sessions, with the following parameters: 0.8 mA current amplitude, 0.1 ms pulse width, 30 Hz frequency, and 0.5 s duration [ 59 , 60 ]. Stimulation was delivered paired with repetitive movements. Data from small pilot and large multicenter RCTs produced similar trends: paired iVNS and motor rehabilitation compared to rehabilitation with sham iVNS was associated with clinically meaningful improvements in participants with moderate to severe arm impairment after ischemic stroke [ 80 ]. In one study, the active iVNS group had two to three times greater improvements than the sham iVNS group, in a population that was at least 9 months post-stroke [ 3 ]. In a population with extremely limited therapeutic options, these results may help fill the void. Overall, the results of these studies are quite exciting, with more than ample evidence of iVNS and its ability to enhance post-stroke recovery. Next steps in this area include assessing whether the model can be used earlier in the post-stroke recovery course and whether these results are applicable in other populations (other types of strokes, other neurological disorders). In 2021, the MicroTransponder Vivistim Paired VNS system (Vivistim system) was approved by the FDA for the treatment of moderate to severe upper extremity motor deficits associated with chronic ischemic stroke. Table 3 VNS for post-stroke motor rehabilitation trials Authors Study/sample size Parameters (active) Implanted/transcutaneous Primary outcome Conclusions Dawson et al. 21 participants with ischemic stroke more than 6 months earlier and moderate to severe upper limb impairment PW: 100 μs Freq: 30 Hz On/off: 500 ms during movement I: 0.8 mA Implanted Change in Fugl-Meyer Assessment–Upper Extremity (FMA-UE) scores in VNS plus rehabilitation group vs. rehabilitation alone In the intention-to-treat analysis, there was no change in the FMA-UE score between groups Dawson et al. RCT of 108 participants with moderate to severe arm weakness, at least 9 months after ischemic stroke PW: 100 μs Freq: 30 Hz On/off: 0.5 s during movement I: 0.8 mA Implanted Change in impairment measured by the Fugl-Meyer Assessment–Upper Extremity (FMA-UE) score The mean FMA-UE score increased significantly more in the VNS group than in controls; a clinically meaningful response on the FMA-UE score was achieved in 47% with VNS vs. 24% in controls Capone et al. 14 patients with either ischemic or hemorrhagic chronic stroke PW: 0.3 ms Freq: 20 Hz On/off: 30 s train/5 min I: varied Transcutaneous Change in upper extremity Fugl-Meyer score between groups receiving robot-assisted therapy with real vs. sham VNS Fugl-Meyer scores were significantly better in the real group than the sham group after 10 days of treatment Redgrave et al. 13 participants at more than 3 months post-ischemic stroke with residual upper limb dysfunction PW: 0.1 ms Freq: 25 Hz On/off: during movement I: Varied Transcutaneous Change in Fugl-Meyer Assessment–Upper Extremity scores after receiving UE rehab + tVNS There was a significant change in the FMA-UE score, with a mean increase per participant of 17.1 points Baig et al. 12 participants at more than 3 months post-ischemic stroke with residual upper limb weakness PW: 0.1 ms Freq: 25 Hz On/off: during movement I: varied Transcutaneous Change in upper limb Fugl-Meyer score from baseline following motor rehab + tVNS for 6 weeks 64% of participants regained some sensation post-intervention, with maximal increase in FMA-UE sensation score seen in the patient with the greatest improvement in motor function Wu et al. 21 subacute ischemia stroke patients with single upper limb motor function impairment PW: 0.3 ms Freq: 20 Hz On/off: 30 s/5 min I: varied Transcutaneous Change in upper extremity Fugl-Meyer Assessment, the Wolf motor function test (WMFT), the Functional Independence Measure (FIM), and Brunnstrom stage from baseline in rehab + active tVNS vs. rehab + sham group after 15 days of treatment The FMA-UE, WMFT, and FIM scores were significantly higher than before treatment, and there was a significantly greater improvement of those measurements in the tVNS group compared with sham-tVNS group Badran et al. 16 post-stroke patients undergoing rehab PW: 500 μs Freq: 25 Hz On/off: 5 s increments during movement I: 1-3 mA Transcutaneous Change in upper limb Fugl-Meyer Assessment, Wolf motor function test Improved upper limb Fugl-Meyer scores compared to sham PW pulse width VNS for post-stroke motor rehabilitation trials PW: 100 μs Freq: 30 Hz On/off: 500 ms during movement I: 0.8 mA PW: 100 μs Freq: 30 Hz On/off: 0.5 s during movement I: 0.8 mA PW: 0.3 ms Freq: 20 Hz On/off: 30 s train/5 min I: varied PW: 0.1 ms Freq: 25 Hz On/off: during movement I: Varied PW: 0.1 ms Freq: 25 Hz On/off: during movement I: varied PW: 0.3 ms Freq: 20 Hz On/off: 30 s/5 min I: varied PW: 500 μs Freq: 25 Hz On/off: 5 s increments during movement I: 1-3 mA PW pulse width Noninvasive VNS has also been studied as a potential therapeutic tool for post-stroke motor recovery. The data in this area are limited but show early promise (see Table  3 ). All the noninvasive studies have utilized taVNS, but their parameters have differed slightly [ 81 , 82 , 83 , 84 ]. Three out of four transcutaneous studies targeted the left cymba concha, while one out of the four targeted the left acoustic meatus. One open-label trial saw 87% of participants ( n  = 13) achieve a clinically relevant increase in mobility as measured by the Fugl Meyer Assessment–Upper Extremity (FMA-UE) [ 82 ]. Three small RCTs have demonstrated similar results, with significant improvements seen in stroke patients receiving active taVNS with physical rehabilitation compared to those receiving sham stimulation with physical rehabilitation [ 81 , 83 , 84 ]. Overall, these studies confirmed the exciting results seen in the implanted cervical VNS population, with active stimulation producing 2–3 times greater improvements than physical rehabilitation alone. Larger trials at multiple centers will be needed before noninvasive options are approved for treatment. A recent pilot study of closed-loop taVNS (called motor-activated auricular vagus nerve stimulation, or MAAVNS) is moving this area of study towards more personalized treatment [ 13 ]. MAAVNS is a closed-loop system designed to improve upper limb function by delivering stimulation precisely in response to movement detected by surface electromyography (EMG) sensors. MAAVNS is individualized to each patient, and the pilot trial demonstrated promising results. The pilot investigated the effects of paired (MAAVNS) and unpaired taVNS on upper extremity motor recovery scores, finding that both groups improved, with a greater effect size in the MAAVNS group. Another trial by the same group investigated the effects of unilateral versus bilateral taVNS on neural activity in chronic stroke patients [ 85 ]. The group measured blood-oxygenation-level-dependent (BOLD) signal propagation in response to ipsilesional versus contralesional versus bilateral versus sham stimulation. The findings suggest that ipsilesional taVNS may be optimal for treatment in post-stroke recovery, as ipsilesional taVNS produced the greatest brain activation in key areas involved in stroke recovery as well as producing greater task-related activation.

As reviewed in the prior sections, VNS is emerging as a promising modality with the ability to treat a wide variety of medical conditions and neuropsychiatric disorders. The field is still young and there are large gaps in the literature. We have highlighted these gaps in each of the preceding sections, but here we give readers a summary and overview of these areas for future improvements. As highlighted in the epilepsy and motor rehabilitation sections, closed-loop VNS systems are beginning to be studied and have made their way into the clinical domain [ 12 , 13 ]. The success of responsive VNS, which sends out trains of stimulation in response to ictal tachycardia, in further decreasing seizure burden compared to standard VNS showcases the promise of closed-loop systems [ 12 ]. Creating personalized VNS paradigms will require greater understanding of the disorders we are treating. VNS combined with other techniques, such as brain imaging or electroencephalography (EEG), will aid in developing personalized treatments. Additionally, noninvasive forms of VNS will be useful in testing, optimizing, and scaling up these novel paradigms. Noninvasive forms of VNS, like taVNS and tcVNS, have created opportunities to more deeply understand how VNS works [ 10 , 11 , 30 ]. Noninvasive VNS has allowed researchers to study the physiologic effects of vagal stimulation in a feasible and cost-friendly manner. Head-to-head trials of noninvasive VNS and iVNS will enable us to understand how these modalities differ. Noninvasive VNS may be useful in predicting response to VNS and prognosis. Additionally, noninvasive VNS will be useful for optimizing parameters. The wearable aspect of noninvasive VNS will allow more clinicians and patients to access this technology and has opened the door for studying VNS in novel settings, including at-home treatment [ 122 ]. Finally, VNS has momentum geared towards future approvals. As highlighted in the previous section, VNS has accumulating evidence in many different medical conditions, including pain disorders, inflammatory disorders, cardiovascular disorders, and gastrointestinal disorders. As we understand more about how VNS works, we will also learn more about the pathophysiology of the disorders we are trying to treat.

Another area that has generated significant interest in VNS research and clinical applications is in the treatment of headaches. Investigations in this domain have mainly used noninvasive VNS modalities (transcutaneous auricular and cervical VNS). One device (marketed as gammaCore) has been cleared by the FDA for acute and preventive treatment of cluster headache and acute treatment of migraine in adults. The gammaCore device uses a form of transcutaneous cervical VNS (tcVNS) that delivers electrical stimulation consisting of five 5000 Hz pulses repeated at a rate of 25 Hz [ 66 ]. This section will broadly overview VNS research into the treatment of headaches and briefly summarize future directions in which the field is heading. Reductions in headache severity and frequency were reported in preliminary case series using invasive iVNS for the treatment of depression or epilepsy [ 86 ]. However, prospective investigations of vagal stimulation with headaches as the primary diagnosis have only used noninvasive VNS methods. The advent of noninvasive VNS has moved this area forward, with many positive RCTs to date. tcVNS, under investigation as gammaCore, has been the most thoroughly studied as an acute treatment for migraine and cluster headache, as well as prophylactic treatment. Early case series of tcVNS for acute migraine showed a significant reduction in duration of pain intensity and headache remission rates [ 87 ]. In one open-label pilot study, 22% of participants were found to be pain-free rate at 2 h, comparable to abortive medications commonly used for migraine (i.e. triptans) [ 88 ]. In a follow-up RCT known as the PRESTO study, the effects of tcVNS on acute migraine were evaluated in a group of 248 patients [ 89 ]. Active stimulation consisted of the parameters described above, delivered for a duration of 120 s, while the sham group received 0.1 Hz biphasic stimulation. The data from this study also showed tcVNS to have equivalent efficacy as medication for acute migraine [ 89 ]. Interestingly, tcVNS was superior to sham at time points 30 and 60 min, but not at 120 min, although the pain-free responder rate was similar to what is seen with triptans and oral nonsteroidal anti-inflammatory drugs (NSAIDs). Studies of tcVNS for cluster headache have also yielded positive results. An open-label pilot found positive results for acute cluster headaches (47% of attacks stopped in 11 min or less) and preventive treatment (reduction in mean attack frequency) [ 88 ]. The open-label trial was followed by two RCTs (ACT1 and ACT2) [ 90 , 91 ]. The ACT1 trial found that active treatment significantly increased the response rate (defined as the proportion of subjects who achieved pain relief) compared to sham stimulation in the episodic cluster headache group, but not the chronic group [ 90 ]. The ACT2 study had similar findings, with benefits seen only in the episodic group and not in those with chronic cluster headache. In contrast, the PREVA study in 2016 found a significant difference between active and sham stimulation in the prevention of chronic cluster headache attacks [ 91 ]. tcVNS has also been studied as a preventive treatment for migraine, but with mixed results. The PREMIUM trial is the largest RCT to date [ 92 ]. Diener’s study consisted of a 12-week double-blind phase followed by a 24-week open-label phase; throughout both phases, a total of 332 patients self-administered stimulation for 120 s twice daily (6–8 h apart). The final analysis, however, did not demonstrate a statistically significant difference between the active and sham groups regarding migraine frequency or reduction. While other forms of noninvasive VNS, like taVNS, have been investigated for different types of headaches, less evidence exists for these modalities. Straube and colleagues led an investigation of 1 Hz taVNS for the prevention of chronic migraine and found that active stimulation was associated with a reduction in number of headache days compared to sham stimulation [ 93 ]. In a separate study, 1 Hz taVNS significantly reduced migraine days, pain intensity, and migraine attack time, while also implicating direct thalamocortical modulation, as measured by change in BOLD functional magnetic resonance imaging (fMRI) signaling, as potentially responsible for the positive therapeutic effects of active taVNS on migraine [ 94 ].

Inevitably, more data will help shape our understanding of how these novel treatments can be made more effective. As in other areas of VNS research, future areas for improvement are in our understanding of how parameters interact with disease state. VNS has an infinite number of parameter combinations, as clinicians can adjust frequency, pulse width, duty cycle, and duration of stimulation. Understanding how these parameter combinations affect treatment outcomes is of utmost importance. Additionally, noninvasive VNS has opened the door for further study of the effects of VNS. This includes translation of the wealth of evidence in animal models to human clinical research, as well as further study of how VNS may be an effective treatment for other medical and neuropsychiatric disorders. Discovery of reliable biomarkers will be crucial in aiding the development of personalized VNS treatments, a revolution that is already occurring in other areas of brain stimulation. In conclusion, the history of VNS spans four decades but has already accomplished a great deal. To date, VNS has gained approvals from the FDA for medication-refractory epilepsy, treatment-resistant depression, obesity, post-stroke motor rehabilitation, and migraine and cluster headache. The development of novel noninvasive forms of VNS promises to further advance the field’s evolution.

The vagus nerve is the tenth cranial nerve and travels from the brainstem, through the neck, and innervates organs throughout the chest and abdomen [ 20 ]. The nerve is present bilaterally and is an essential component of the parasympathetic nervous system. Around 80% of the nerve’s fibers are afferent, carrying sensory information from the viscera to the brain. The various functions of the vagus can be targeted using electrical stimulation, eliciting central and peripheral effects [ 20 ]. Due to its widespread projections and various functions, the mechanism of therapeutic action of VNS for any given disorder is likely multifactorial, with multiple mechanisms working synergistically. For example, major depression is associated with alterations in the brain and periphery, including immune and endocrine dysfunction [ 21 ]. VNS has been shown to modulate neural activity as well as modulate the neuroendocrine–immune axis and function [ 5 ]. Thus, it is likely that there are multiple mechanisms by which VNS exerts its effects in depression alone. This section will give readers an overview of all the potential mechanisms of VNS. Implanted VNS and noninvasive VNS have similar mechanisms, although the difference between the two modalities is not well understood as there has never been a head-to-head trial comparing their effects. VNS produces anticonvulsant effects, first discovered by Zabara in 1985 [ 18 ]. While the antiepileptic mechanisms of VNS are not fully understood, evidence suggests that norepinephrine plays a crucial role [ 22 ]. The vagus nerve projects through the locus coeruleus, or the major noradrenergic nucleus in the brain. In a rodent model, researchers ablated the locus coeruleus and found that VNS lost its anti-seizure effects [ 22 ]. Other mechanisms likely contribute to the anticonvulsant effects of VNS, but more data are needed. As highlighted in the previous paragraph, VNS modulates the monoamine system (e.g., serotonin, norepinephrine). VNS has been shown to increase the concentration of monoamines found in the CSF [ 23 ]. In rodent models, VNS increases the firing rate of norepinephrine and serotonin neurons [ 24 ]. This mechanism likely plays a key role in the antidepressant effect of VNS. Additionally, neuroimaging studies have demonstrated that VNS modulates key areas of the brain involved in regulating mood, including the prefrontal cortex and limbic system [ 5 ]. Animal models have also been helpful in elucidating the neural plasticity-enhancing effects of VNS. VNS increases the expression of brain-derived neurotrophic factor (BDNF) and synaptic spine density in the hippocampus [ 25 ]. This mechanism likely plays a role in the mechanism of VNS for multiple disorders, including depression and post-stroke motor rehabilitation [ 26 ]. When paired with a motor task, VNS reorganizes the primary motor cortex and increases the representation of behavioral tasks in associated brain regions [ 27 ]. The same concept has been demonstrated in non-motor paradigms, where VNS paired with an auditory tone reorganizes and increases plasticity within the primary auditory cortex [ 28 ]. VNS has also been shown to modulate immune function [ 29 ]. It decreases immune markers and inflammatory cytokines [ 5 ]. This has been shown to be correlated with the therapeutic effects of VNS for post-traumatic stress disorder (PTSD) [ 30 ]. VNS also exerts cardiovascular effects, closely related to modulation of the autonomic system. Acute trains of VNS decrease heart rate and other sympathetic markers [ 11 ]. VNS has also been shown to decrease the heart rate during the stress response [ 31 ]. Additionally, it has been shown to reduce sympathetic activity and improve baroreflex control [ 32 , 33 ], which are likely mechanisms in treating conditions that are characterized by sympathetic overactivity. The gastrointestinal effects of VNS likely work through a concert of multiple systems along the brain–gut axis. Specifically, the gastrointestinal effects of VNS work closely in concert with the cholinergic anti-inflammatory pathway [ 34 ].

VNS is being studied as a potential therapeutic for a variety of other conditions, including pain, inflammatory disorders, cardiovascular diseases, and gastrointestinal disorders. These conditions have been predominately studied using noninvasive VNS. This section will briefly overview exploratory work that has been conducted in each category. The analgesic effects of VNS have been described since the modality’s inception. The interplay between various VNS parameters and their behavioral effects remains to be determined. Here, we give a brief overview of VNS for pain studies. VNS for headache and migraine studies is discussed in the headache section above. VNS was found to be effective for a variety of conditions associated with chronic pain. In addition to the headache and migraine studies, the antinociceptive effects of VNS have been demonstrated for conditions such as fibromyalgia, pancreatitis, irritable bowel syndrome (IBS), esophageal pain, and polymyalgia rheumatica [ 7 ]. The advent of noninvasive VNS has allowed more feasible and less costly study of VNS and its effects on these pain-related disorders. The Cerbomed Nemos taVNS device (tVNS Technologies, Erlangen, Germany) received European certification for chronic pain in 2012. However, several iVNS trials have demonstrated analgesic effects. For example, an interesting trial of iVNS (for coronary artery disease [CAD]) abolished angina at rest, as well as reducing heart rate and blood pressure [ 95 ]. taVNS successfully decreased pain and fatigue in a group of patients ( N  = 18, randomized 2:1 to receive active) with systemic lupus erythematosus (SLE) [ 96 ]. Bellocchi found similar results for SLE-related pain, using taVNS as an adjunct to usual treatment [ 97 ]. The group also found downregulated interleukin-6 levels in the active group compared to sham. Interestingly, there was no significant effect on quality-of-life scales or heart rate variability, which have been seen in other VNS studies (e.g., for major depression and epilepsy). taVNS has also been shown to relieve abdominal pain and constipation associated with IBS [ 98 ]. Kovacic et al. found that impaired cardiac vagal regulation (measured by vagal efficiency) predicted pain improvement associated with taVNS [ 99 ]. tcVNS was investigated as a potential therapeutic for pain associated with chronic pancreatitis, but no difference was found in pain scores between active and sham treatment [ 100 ]. In a single-blind sham-controlled study, taVNS combined with deep slow breathing enhanced gastroduodenal motility and antral contractions but did not demonstrate any significant effect on pain thresholds [ 101 ]. Napadow and colleagues described a modified form of taVNS called respiratory-gated auricular vagal afferent nerve stimulation (RAVANS), which they used to study chronic pelvic pain due to endometriosis. They demonstrated reductions in evoked pain intensity and temporal summation of mechanical pain, as well as anxiety [ 102 ]. Vagal modulation of pain has also been studied in healthy subjects. Reports indicate that taVNS may increase mechanical and pressure pain thresholds as well as mechanical pain sensitivity [ 103 ]. Alt and colleagues demonstrated reductions in pain unpleasantness ratings due to taVNS [ 104 ]. taVNS has also been shown to decrease acid-induced esophageal hypersensitivity [ 105 ]. However, some trials did not show any effect of taVNS on pain in healthy adults [ 31 ]. VNS modulation of the neuroimmune system has been well studied in animal models, but less so in humans. VNS has been shown to decrease inflammation associated with irritable bowel disease, as well as inflammation associated with stroke, traumatic brain injury (TBI), and depression [ 106 ]. Most studies to date in this arena have been preclinical animal models, but noninvasive VNS represents an opportunity for further translation of these models into the clinical domain. In a study of iVNS for Crohn’s disease, five out of nine patients experienced remission of symptoms, as well as decreased levels of inflammatory markers (C-reactive protein and calprotectin) [ 107 ]. Another pilot trial had similar, although less impressive, results. Four out of 16 patients implanted with iVNS experienced remission of Crohn’s symptoms [ 108 ]. Additionally, iVNS decreases cytokine production and attenuates disease severity in rheumatoid arthritis. Promising results have been seen in the use of taVNS in small cohorts of patients with various inflammatory disorders. In a study of patients with systemic lupus erythematosus (SLE), plasma levels of Substance P were significantly reduced after 5 days of taVNS [ 96 ]. taVNS reduced pain scores, as well as downregulating IL-6, in patients with systemic sclerosis [ 97 ]. Additionally, patients with sepsis showed a reduction in inflammatory cytokines after five consecutive days of taVNS [ 109 ]. taVNS decreased levels of inflammatory cytokines associated with myocardial ischemia and reperfusion, atrial arrhythmias, and heart failure [ 110 ]. An interesting pilot trial of taVNS for symptoms associated with long COVID demonstrated decreases in anxiety and fatigue, which was hypothesized to be due to a decrease in inflammation associated with the development of long COVID [ 111 ]. Several large trials have investigated the effectiveness of invasive VNS for heart failure [ 112 , 113 , 114 ]. Unfortunately, these trials did not report significant differences in outcomes between active and sham groups. Premchand and colleagues demonstrated improved left ventricular ejection fraction by 4.5% in patients 10 weeks after the implantation of iVNS in a study of 60 patients with heart failure [ 115 ]. Zamotrinsky and colleagues found that transcutaneous VNS treatment abolished angina at rest and reduced heart rate and blood pressure [ 116 ]. They also found improvements in left ventricular ejection fraction. Yu found similar results—taVNS reduced ventricular arrhythmia and improved left ventricular ejection fraction [ 117 ]. Additionally, VNS has been shown to work through other cardiovascular mechanisms, including lowering sympathetic activity and improving baroreflex control. [ 32 , 33 ]. These could be potential mechanisms for the treatment of conditions characterized by sympathetic overactivity, like hypertension and chronic kidney disease. Additionally, cardiovascular mechanisms, including lowering of sympathetic activity, may play a role in the psychotherapeutic effects of VNS. VNS has been studied in a variety of gastrointestinal disorders. Implanted VNS is FDA-approved for obesity but has early evidence for other gastrointestinal disease. It is likely effective for gastrointestinal disease through a combination of its parasympathetic (autonomic afferent and efferent) functions and its ability to modulate the neuroimmune and neuroendocrine systems. All the early studies for gastrointestinal disease have used noninvasive VNS. After 4 weeks of taVNS, patients with constipation-predominant irritable bowel syndrome (IBS-C) had decreased abdominal pain and increased frequency of complete bowel movements [ 98 ]. Patients with Parkinson’s disease who received taVNS had improved scores on the Gastrointestinal Symptom Rating Scale [ 118 ]. Zhang and colleagues investigated the effectiveness of a similar treatment, transcutaneous electrical acustimulation (TEA), on gastric motility and gastroesophageal reflux disease [ 119 ]. They found that TEA had positive effects on reflux-related symptoms and motility. In a trial using tcVNS (gammaCore), 10 out of 23 patients with drug-resistant gastroparesis saw a response (measured as improvements in nausea/vomiting, postprandial fullness/early satiety, and bloating) [ 120 ]. Another pilot trial had similar results, with 40% of patients reporting improvement in symptoms related to gastroparesis, including accelerated gastric emptying [ 121 ].

Vagus nerve stimulation (VNS) is an exciting modality that has already demonstrated potential to treat a variety of medical and neuropsychiatric disorders. Historically, VNS was found to be an effective treatment for epilepsy, and later, developed as a treatment for major depression [ 1 ]. Over the past several decades, VNS has gained approvals in other areas, including obesity, post-stroke motor rehabilitation, and migraine [ 1 ]. With increasing interest in VNS for myriad conditions, ongoing investigations will likely lead to additional approvals. This review serves to give readers an overview of the science leading to the current clinical indications of VNS, along with the current state of research that may guide the field to future approved indications. VNS is in prime position to gain additional approvals for neuropsychiatric disorders over the coming years. As outlined in the following sections, we will discuss the current state of VNS for approved indications as well as highlight future areas for growth. There is a plethora of evidence supporting the potential for VNS to treat a variety of disorders, including neurological and psychiatric disorders [ 2 – 4 ], inflammatory and immune disorders [ 5 , 6 ], pain-related disorders [ 7 ], cardiovascular diseases [ 8 ], and other diseases related to autonomic dysfunction [ 9 ]. The vagus nerve has many functions and innervates various end-organs throughout the chest and abdomen, which highlights its therapeutic potential for a variety of conditions [ 1 ]. But access to VNS has been limited over the first few decades of use due to the invasive and expensive nature of implanted VNS. The recent development of noninvasive forms of VNS have opened the door for scientists to study the effects of vagal stimulation more closely [ 10 , 11 ]. Additionally, closed-loop VNS systems are beginning to make their way into research and clinical domains [ 12 , 13 ]. In this era of personalized medicine, researchers are studying ways to make VNS treatments individualized and unique. There is still much for us to learn about this modality. In the following sections, we will further highlight and explore the current gaps in the literature, in addition to the ones already described here.