Multimodal Neural Assessment for Predicting Post-Stroke Dysphagia: Identifying Critical Risk Factors Using Combined Indicators.

BACKGROUND Post-stroke dysphagia (PSD) severely impairs patients' quality of life, yet current diagnostic methods lack efficacy in early detection and prediction. This retrospective study of 300 stroke patients aimed to evaluate risk factors in patients with and without post-stroke dysphagia using functional near-infrared spectroscopy (fNIRS), transcranial magnetic stimulation (TMS), and surface electromyography (sEMG). MATERIAL AND METHODS A total of 300 stroke patients (201 with PSD and 99 without, diagnosed via video fluoroscopic swallowing study or flexible endoscopic evaluation of swallowing) were included. Statistical power analysis determined the sample size, logistic regression identified independent risk factors, and receiver operating characteristic (ROC) curves evaluated the predictive accuracy of combined fNIRS-TMS-sEMG indicators. RESULTS Independent risk factors for PSD included age >65 years, stroke location, Modified Barthel Index score <45, National Institutes of Health Stroke Scale score, clustering coefficient (Cp), peak amplitude (PA), and motor-evoked potentials (MEPs) (all P<0.1). The combination of Cp+PA+ipsilateral MEP+contralateral MEP showed the highest predictive accuracy (area under the curve [AUC]=0.985, P<0.001, 95% CI 0.974-0.995), with 92.0% sensitivity and 96.0% specificity, and an optimal cutoff value of 0.737. CONCLUSIONS This study identified key PSD risk factors. The combination of fNIRS, TMS, and sEMG provides a highly accurate approach for early PSD prediction. Future research should compare this multimodal method with traditional diagnostics to confirm its clinical value.