Cognitive and behavioral effects of prefrontal rTMS in post-stroke cognitive impairment: a randomized sham-controlled trial

Repetitive transcranial magnetic stimulation (rTMS) has shown promise in improving cognitive function in post-stroke cognitive impairment (PSCI), yet its effects under varying cognitive demands and associated neurophysiological changes remain incompletely characterized. In this randomized, double-blind, sham-controlled trial, 75 patients with PSCI were randomized, of whom 68 (rTMS: n = 34; sham: n = 34) were included in the modified intention-to-treat clinical analysis. Participants received active or sham rTMS over the left dorsolateral prefrontal cortex for 2 weeks. The principal clinical outcome analyzed in this report was change in Montreal Cognitive Assessment (MoCA). Secondary outcomes included Hamilton Depression Rating Scale (HAMD), Mini-Mental State Examination (MMSE), and performance during a working memory (n-back) task. Exploratory EEG and fNIRS analyses were performed to evaluate neurophysiological changes associated with rTMS. Compared with sham stimulation, active rTMS was associated with greater improvement in global cognitive function (MoCA group × time interaction: β = 1.76, 95% CI 0.69 to 2.84, p = 0.001), with significant improvements also observed in MMSE and depressive symptoms. Condition-specific analyses showed improvements in 1-back reaction time and accuracy but not in the 0-back condition; however, group × time × condition interactions were not statistically significant. No formally tested EEG spectral-power, EEG PLV-based functional-connectivity, or task-evoked fNIRS HbO comparison remained statistically significant after FDR correction. High-frequency rTMS over the left DLPFC was associated with short-term cognitive improvement in patients with PSCI. Behavioral improvements were observed in the 1-back condition, although formal interaction analyses did not establish differential treatment effects between task conditions. Neurophysiological findings remained exploratory and require validation in larger adequately powered studies.