Motor evoked potentials in predicting recovery from upper extremity paralysis after acute stroke

Objective: The use of motor evoked potentials (MEPs) in predicting recovery after stroke still appears to be somehow equivocal. We assessed the prognostic value of MEPs with respect to arm and hand motor recovery in acute stroke patients. Methods: This cohort study included 43 consecutive acute stroke patients with complete paralysis of the upper extremity. MEPs of the abductor digiti minimi muscle (ADM) and the biceps brachii muscle (BB) were obtained within 10 days after stroke onset. The upper limb subset of the Fugl-Meyer Motor Assessment was used to evaluate the motor performance at regular intervals until 6 months after stroke. Results: The follow-up was complete in 40 patients (2 patients died and 1 patient had a recurrent stroke); 14 patients showed motor recovery of the arm and their mean 26-week arm motor score was 17.93 (range 3-30, SD 11.68); hand motor recovery occurred in 11 patients and their mean 26-week hand motor score was 11.09 (range 4-14, SD 4.10). Stepwise logistic regression revealed prognostic models for both arm and hand motor recovery based on BB MEPs (odds ratio 7.69, confidence interval, CI, 1.16-50.95) and ADM MEPs (odds ratio 16.20, CI 2.51-104.40), respectively. Conclusions: The predictive significance of MEPs with respect to motor recovery of the upper extremity was obvious in our homogeneous sample of patients. This agrees with the paradigm that motor recovery after infarction is strongly dependent on a critical residual sparing of the corticospinal function. In this context, the test properties of MEPs in predicting motor recovery are discussed. The added value of MEPs with respect to motor recovery of the upper extremity should be regarded as established for patients with initial paralysis, especially since clinical examination alone lacks the possibility to detect the potential for motor recovery in these cases. Copyright (C) 2003 S. Karger AG, Basel.