Predicting efficacy of combined functional electrical stimulation combined with mirror therapy in post-stroke lower limb dysfunction: development and validation of a nomogram model.

Objective: To investigate the predictive factors of the efficacy of conventional therapy combined with functional electrical stimulation combined with mirror therapy (FES-MT) in the treatment of lower limb dysfunction after cerebral infarction, and to construct and validate a nomogram prediction model to assist in clinical precision treatment. Methods: A total of 212 patients with lower limb dysfunction after cerebral infarction, who received the aforementioned treatment, were enrolled retrospectively. Based on therapeutic efficacy, they were divided into an effective group and an ineffective group. Relevant clinical indicators were collected. Patients were randomly assigned in a 7:3 ratio to a training set (n = 148) and a validation set (n = 64). In the training set, multivariate logistic regression analysis was performed to identify independent influencing factors, and a nomogram prediction model was constructed. The predictive performance of the model was evaluated using calibration curves, the concordance index (C-index), and decision curve analysis. Results: Among the 212 patients, 142 were classified as effective and 70 as ineffective. Multivariate logistic regression analysis showed that longer time from onset to treatment, higher fasting plasma glucose, higher homocysteine, larger infarct volume, higher National Institutes of Health Stroke Scale score, and lower pre-treatment Fugl-Meyer score (lower limb) were independent risk factors for ineffective treatment. The constructed nomogram model exhibited C-index values of 0.890 and 0.872 in the training and validation sets, respectively, with mean absolute errors of 0.124 and 0.114. In the Hosmer-Lemeshow test, the χ2 values were 9.907 (p = 0.271) for the training set and 7.194 (p = 0.515) for the validation set. Receiver operating characteristic curve analysis showed that the area under the curve values were 0.890 (95% CI: 0.822-0.957) and 0.872 (95% CI: 0.747-0.998) for the training and validation sets, respectively, with combined sensitivity and specificity of 0.847, 0.875 and 0.812, 0.750. Conclusion: The nomogram prediction model, based on the identified influencing factors, demonstrates excellent predictive performance for the clinical efficacy of conventional therapy combined with FES-MT in patients with lower limb dysfunction after cerebral infarction. This model facilitates clinical outcome prediction and aids in formulating individualized treatment strategies.