Immediate changes during dysesthesia-matched transcutaneous electrical nerve stimulation in refractory neuropathic pain: a retrospective observational case series

Background Dysesthesia—such as tingling and numbness—remains refractory to pharmacological treatment and poses therapeutic challenges within neuropathic pain. Dysesthesia-matched transcutaneous electrical nerve stimulation (DM-TENS) is a perceptually guided neuromodulation approach in which stimulation parameters are iteratively adjusted to synchronize with the perceived temporal patterns and intensity of the patient’s abnormal sensory experience. However, immediate responses to DM-TENS across heterogeneous neuropathic pain conditions, as well as the feasibility of both direct and indirect (remote-site) stimulation approaches have not been described. Methods This retrospective observational case series included consecutive patients with refractory neuropathic pain referred from the pain clinic to the rehabilitation department at our university hospital between December 2024 and August 2025. Dysesthesia intensity was assessed before and during stimulation using the Numerical Rating Scale (NRS), and pain quality was evaluated using the Short-Form McGill Pain Questionnaire-2 (SF-MPQ-2) as an exploratory measure. Within-session changes were examined using session-level linear mixed-effects models, with aggregated non-parametric analyses performed as sensitivity analyses. As an exploratory analysis, within-patient consistency of stimulation parameters (frequency, pulse width, and intensity) across repeated sessions was evaluated using intraclass correlation coefficients (ICCs). Results DM-TENS was well tolerated in all participants, with no stimulation-related adverse events requiring session termination. Fifteen patients underwent 66 DM-TENS sessions, of which 64 provided NRS data. A session-level linear mixed-effects model showed significantly lower NRS scores during stimulation ( p < 1 × 10⁻ 8 ). Aggregated sensitivity analyses showed a median NRS change of − 2.0 (interquartile range − 2.75 to − 1.43). Indirect stimulation was applied only in trigeminal neuropathic presentations; lower within-session symptom scores were observed during palm stimulation, and these findings should be considered exploratory. SF-MPQ-2 total scores were also significantly lower during stimulation ( p = 0.002). Stimulation parameters showed good to excellent within-patient consistency across repeated sessions (ICC range 0.78–0.85). Conclusions In this retrospective observational case series, lower within-session symptom scores were observed during DM-TENS across heterogeneous neuropathic pain conditions, including trigeminal, spinal, and peripheral presentations. These findings raise the hypothesis that individualized, perceptually guided adjustment of stimulation parameters warrants prospective evaluation.