Transcranial magnetic stimulation combined with functional near-infrared spectroscopy to elucidate the neurophysiological mechanisms of post-stroke hemiplegia: a systematic review.

Patients with hemiplegia after stroke often suffer from motor dysfunction, and their recovery is frequently limited by the imbalanced interhemispheric competition and abnormal neural network reorganization. The combined application of transcranial magnetic stimulation (TMS) and functional near-infrared spectroscopy (fNIRS) has provided a novel closed-loop neuromodulation strategy for the assessment and intervention of post-stroke motor function. This systematic review aims to systematically synthesize clinical research evidence regarding the combined use of TMS and fNIRS in post-stroke motor function recovery, as well as to analyze its neurophysiological mechanisms, clinical efficacy, and methodological quality. By searching databases including PubMed, Web of Science, Embase and Cochrane Library, a total of 9 studies were finally included. The results showed that the TMS-fNIRS combined protocol was safe and feasible, with six of the nine interventional studies reporting statistically significant improvements in motor function (e.g., Fugl-Meyer Assessment scores), while the remaining three did not report clinical outcomes. At the neurophysiological level, effective TMS interventions (such as intermittent theta-burst stimulation (iTBS) or high-frequency repetitive transcranial magnetic stimulation (rTMS) could enhance the activation of the ipsilesional primary motor cortex, facilitate the restoration of interhemispheric balance (with the laterality index shifting toward the ipsilesional side), and improve local and interhemispheric functional connectivity as well as brain network efficiency. Two-thirds of the studies (6/9) found that changes in neurophysiological indicators were significantly correlated with improvements in clinical functions, providing preliminary, hypothesis-generating correlational support for the underlying intervention mechanisms. However, existing studies have limitations including small sample sizes, high protocol heterogeneity, and uneven methodological quality (especially the high risk of bias in non-randomized studies), resulting in a moderate to low level of evidence strength. In conclusion, the combined TMS-fNIRS technology demonstrates the potential to advance stroke rehabilitation toward individualization and closed-loop practice, but current evidence is still in the early stage. Future research needs to conduct large-scale, standardized, algorithm-driven clinical trials to achieve the transition from "proof-of-concept" to "precision therapy".