Article

This study investigated the clinical predictive value of the
Fugl-Meyer Assessment (FMA) arm score and the upper limb activity assessed by
accelerometers in patients with hemiparesis after acute stroke. DESIGN: The
prospective cohort (n = 129) was recruited from a general hospital; activity
variables and FMA score at intake were related to the FMA, the modified Rankin
Scale, and rehabilitation status after 3 mos of follow-up. The prediction model
was based on binary logistic regression. RESULTS: Although the FMA score at
intake has the best overall predictive value for all three outcome measures
(FMA3, 87.6%; modified Rankin Scale, 85.3%; RS, 73.6%), the activity of the
impaired arm as assessed by the accelerometer has the best predictive value to
determine patients who are at risk for continued disability (modified Rankin
Scale score 1, 95.1%). The most difficult outcome measure for prediction is the
rehabilitation status; specifically, the patients who went home are predicted
imprecisely. The ratio variable is the least accurate predictor of all tested
variables. CONCLUSIONS: The FMA arm score at intake is the best predictor for arm
recovery and general disability. The activity of the impaired arm is an excellent
predictor for prolonged disability and is an alternative to the FMA score when it
is impossible to score the FMA in the acute phase of stroke.

Langue : ANGLAIS