BACKGROUND: Our understanding of osteoarthritis (OA) has evolved from a degenerative disease to one in which low-grade, chronic inflammation plays a central role. In addition, evidence suggests that OA is accompanied by both peripheral and central nervous system sensitization that can cause pain. It has been demonstrated that transcutaneous vagus nerve stimulation (tVNS) can relieve pain, inflammation, and central sensitization in other conditions including fibromyalgia, pelvic pain, and headaches. We aimed to assess the efficacy and safety of tVNS on nociceptive pain, central sensitization, and physical function in knee OA. METHODS: In this 12-week study, we stimulated the auricular branch of the vagus nerve with an auricular electrode connected to a transcutaneous electrical nerve stimulation device once a day for 3 days each week for 12 weeks. A total of 68 patients with chronic knee OA were randomly assigned to the active and sham groups (34 patients in each group). We used a variety of outcome measures, including the visual analog scale (VAS), pressure pain threshold (PPT), knee injury and osteoarthritis outcome score (KOOS), PainDETECT (PD-Q) and Douleur Neuropathique 4 (DN4) questionnaires. Outcome measures were recorded at baseline, At the end of the stimulation period, and then after 4 weeks. RESULTS: In the active group, compared to baseline, there was a significant improvement in VAS scores between the first and second follow-up visits (P < 0.001). A significant improvement in PPT was seen in the right knee, left knee, and right elbow in active tVNS, and this improvement persisted for four weeks post-intervention. Meanwhile, in the sham group, right knee PPT was improved but not maintained. There were statistically significant improvements in the PD-Q and DN4 scores in the active tVNS group (P < 0.001), whereas in the sham group, DN4 questionnaire did not show any improvement. In terms of functional outcomes, the improvement in KOOS was significant only in the active group (31.44 ± 18.49, P < 0.001). No serious adverse events were observed. CONCLUSION: There is preliminary evidence to support the benefits of tVNS in OA, suggesting that neuromodulation can be used as an adjunct to existing pharmacological and non-pharmacological treatments. TRIAL REGISTRATION: The study was registered on ClinicalTrials.gov (NCT05387135) on 24/05/2022.