Predicting hand motor recovery in severe stroke: the role of motor evoked potentials in relation to early clinical assessment.

OBJECTIVE: The primary aim of this study was to compare the predictive value of motor evoked potentials (MEPs) and early clinical assessment with regard to long-term hand motor recovery in patients with profound hemiplegia after stroke. METHODS: The sample was an inception cohort of 39 stroke patients with an acute, ischemic, supratentorial stroke and an initial upper-extremity paralysis admitted to an academic hospital. Hand motor function recovery was defined at 26 weeks poststroke as a Fugl-Meyer Motor Assessment (FMA) hand score>3 points. The following prognostic factors were compared at week 1 and week 3 poststroke: motor functions as assessed by the FMA upper-extremity and lower-extremity subscores, and the presence of an MEP in the abductor digiti minimi and biceps brachii muscle. RESULTS: Both the presence of an abductor digiti minimi-MEP and any motor recovery in the FMA upper-extremity subscore showed a positive predictive value of 1.00 at weeks 1 and 3. The FMA lower-extremity subscore showed the best negative predictive value (0.90; 95% CI 0.78-1.00 at week 1 and 0.95; 95% CI 0.87-1.00 at week 3). CONCLUSIONS: In stroke patients with an initial paralysis of the upper extremity the presence or absence of an MEP has similar predictive value compared with early clinical assessment with regard to long-term hand motor recovery