Systematic Review | Clinical Evidence and Mechanistic Advances in Vagus Nerve Stimulation for Rheumatic Diseases

 Pain, fatigue, and recurrent inflammation are long-term problems for many people with rheumatic diseases. Drug treatment has improved substantially, yet some patients still do not achieve adequate disease control. Researchers have therefore begun to explore a complementary strategy in which the nervous system is regulated to influence immune activity linked to inflammation.

In 2026, Anna Wysokińska and colleagues published a review in Rheumatology International that examined the mechanisms and clinical evidence for vagus nerve stimulation (VNS) in rheumatic diseases. The authors included 15 original human studies covering rheumatoid arthritis, systemic lupus erythematosus, Sjögren's syndrome, spondyloarthritis, osteoarthritis, fibromyalgia, and other conditions.

VNS has shown potential benefits for inflammation, pain, fatigue, and disease activity. Findings have varied across diseases and studies, however, and there is still no common standard for stimulation methods, treatment parameters, or follow-up duration.

Background：Neuroimmune Pathways

Autonomic nervous system (ANS) dysfunction is common in rheumatic diseases and often appears as increased sympathetic activity alongside reduced parasympathetic regulation mediated by the vagus nerve. The review notes that this imbalance accompanies chronic inflammation, but current evidence cannot determine whether autonomic dysfunction is a consequence of disease or contributes to its onset and persistence.

Approximately 80% of vagal fibers are afferent and 20% are efferent. After afferent signals enter the brainstem, structures such as the nucleus tractus solitarius can influence the hypothalamic-pituitary-adrenal axis, which helps regulate stress responses and cortisol. The efferent pathway is linked to the cholinergic anti-inflammatory pathway. When acetylcholine binds to α7 nicotinic acetylcholine receptors on macrophages, it can suppress the production of pro-inflammatory cytokines such as tumor necrosis factor α. Regulation involving the spleen also includes the splenic nerve and acetylcholine-releasing T lymphocytes, with several components acting together across the pathway.

Heart rate variability (HRV) is often used to assess autonomic regulation. Lower HRV generally suggests reduced vagal tone or greater sympathetic dominance. HRV is a physiological measure, however, and cannot stand in for disease activity or clinical efficacy.

Methods：Evidence Search and Classification

The authors searched PubMed/MEDLINE from January through August 22, 2026, and conducted an additional search of the Directory of Open Access Journals. They placed no restriction on publication year. The search identified 143 records, of which 140 remained after limiting the results to English-language publications. Screening of titles, abstracts, and full texts left 15 original clinical studies for inclusion. Follow-up or extension reports from the same patient cohort were not counted as separate studies.

Eligible studies included randomized and nonrandomized clinical trials, sham-controlled studies, pilot studies, proof-of-concept studies, feasibility studies, and single-arm studies. The authors excluded animal and in vitro experiments, reviews, meta-analyses, study protocols, commentaries, case reports, and studies that did not address either rheumatic disease or VNS.

The studies did not use a single stimulation technique. Implanted VNS places an electrode on the cervical vagus nerve. Transcutaneous cervical stimulation delivers stimulation through the skin of the neck. Transcutaneous auricular vagus nerve stimulation (taVNS) targets auricular branches of the vagus nerve. The review also included manual auricular acupuncture. Results obtained with different devices, stimulation sites, and doses should therefore not be pooled or interpreted as equivalent.

Results：Variation in Clinical Outcomes

The review found that the clinical evidence was most concentrated in rheumatoid arthritis (RA). One implanted VNS study of 17 participants used 10 Hz stimulation and reported a decrease in the 28-joint Disease Activity Score with C-reactive protein (DAS28-CRP), from 6.05 ± 0.18 at baseline to 4.16 ± 0.39 on day 42. Tumor necrosis factor levels also fell from 2,900 ± 566 pg/mL to 1,776 ± 342 pg/mL. In another study of 30 participants receiving taVNS, 53% met the American College of Rheumatology criterion for 20% improvement after 12 weeks. The proportions meeting the 50% and 70% improvement criteria were 33% and 17%, respectively.

The largest implanted VNS study included in the review enrolled 242 participants. At 3 months, 35.2% of the stimulation group and 24.2% of the control group met the American College of Rheumatology criterion for 20% improvement. The stratified adjusted between-group difference was 11.8 percentage points, with a 95% confidence interval of 0.6 to 23.1 and P = 0.0209. These findings came from left-sided cervical implantation with stimulation at 10 Hz. They do not establish that transcutaneous stimulation at the ear or neck would produce the same effect.

The findings for auricular stimulation did not point in a single direction. In an RA study of 113 participants, the change in DAS28-CRP at week 12 was −0.95 ± 0.16 in the stimulation group and −0.66 ± 0.16 in the sham group. Changes in the Health Assessment Questionnaire Disability Index were −0.19 ± 0.06 and −0.02 ± 0.06, respectively. The review table did not report confidence intervals or P values for either between-group difference, so the size of the changes alone does not confirm efficacy.

Evidence from other diseases likewise included both positive and negative findings. A study of 18 participants with systemic lupus erythematosus reported improvements in pain, fatigue, and joint measures. Studies in Sjögren's syndrome found lower fatigue and symptom scores, but no significant between-group difference in HRV. A study of erosive hand osteoarthritis enrolled 142 participants. At week 12, the adjusted between-group difference in the change in hand pain was −10.0 mm, with a 95% confidence interval of −23.0 to 2.0 mm and P = 0.22, indicating no significant between-group difference.

Mechanistic Interpretation：Changes in Measures and Their Limits

The results show that symptoms, inflammatory markers, and autonomic measures do not necessarily change together. Some studies found less pain or fatigue without a change in HRV. Others recorded an increase in HRV, but whether clinical symptoms improved and remained better was unclear. HRV can help researchers track autonomic responses, but it cannot currently serve as a surrogate measure of clinical efficacy.

The authors suggest that low-frequency stimulation at about 10 Hz may be more favorable for anti-inflammatory effects, whereas higher frequencies of 20 to 25 Hz may exert a greater influence on afferent pathways and may be associated with improvements in pain and fatigue. This interpretation comes from comparisons across separate studies whose participants, diseases, stimulation sites, and devices also differed. The review did not identify direct comparisons among stimulation parameters, so no single optimal frequency can yet be established.

Most studies in the review had small samples, short stimulation and follow-up periods, and markedly different protocols. The outcomes assessed were also inconsistent. Beyond inflammatory markers, pain, and disease activity, few studies systematically examined daily functioning, comorbid symptoms, analgesic use, or health-care utilization. Potential adverse events also require more complete reporting.

Future Directions：A Standardized Path to Validation

Future research could use larger samples and multicenter, sham-controlled trials conducted across different regions, with consistent clinical endpoints and longer follow-up. Separate comparisons of stimulation frequency, intensity, site, and treatment duration are needed to identify which parameters are associated with changes in inflammation, pain, or fatigue. Studies could also record HRV, inflammatory markers, daily functioning, medication use, and adverse events in parallel. This would test whether physiological changes predict clinical benefit and make findings easier to compare across studies.

At-Home taVNS and ZenoWell Luna Plus

The studies reviewed here used different forms of vagus nerve stimulation, including implanted cervical VNS and transcutaneous auricular stimulation, so their results cannot be applied interchangeably to consumer devices. ZenoWell Luna Plus is a non-invasive, ear-based taVNS wellness device designed for relaxation and stress-management routines. It should not be interpreted as a treatment for rheumatoid arthritis, lupus, osteoarthritis, or other rheumatic diseases. For people interested in at-home taVNS, the current evidence is better viewed as a reason for continued research than as proof of disease-specific benefit.

References

Wysokińska A, Jaszczyk B, Sadura-Sieklucka T. Vagus nerve stimulation in rheumatic diseases: linking autonomic nervous system regulation, inflammation, and clinical outcomes. Rheumatology International. 2026;46:260.