Prediction of upper limb recovery in the acute phase of cerebrovascular disease: evaluation of "functional hand" using the manual function test.

BACKGROUND

Prediction of upper limb function recovery in the acute phase of cerebrovascular disease can help clarify goal setting in rehabilitation and subsequently shorten hospital stay. The present study aimed to develop regression equations that can be used to predict the Manual Function Test (MFT) score 3 weeks after onset and to determine the optimal cutoff MFT score for the identification of "Functional Hand."

METHODS

In all, 190 patients with cerebrovascular disease were included in this study. The baseline survey performed within 1 week after onset assessed sociodemographic profiles, medical profiles, and acute symptoms. MFT was performed to determine the cutoff score to indicate Functional Hand. We used stepwise multiple regression analysis to establish the prediction equations with the best fit for the MFT score 3 weeks after onset. In addition, the sensitivity and specificity of the MFT as an indicator of Functional Hand with cutoff values were determined.

RESULTS

The multiple regression analysis showed that the following factors had a significant influence on the MFT: Brunnstrom recovery stage, cognitive function, range of motion, age, and sensation. The area under the curve was .93 for the MFT score as an indicator of Functional Hand. The cutoff MFT score to identify Functional Hand was 22/21 points, with a sensitivity and specificity of 91.1% and 82.1%, respectively.

CONCLUSIONS

Our findings helped develop regression equations that can be used to predict the MFT score 3 weeks after onset of cerebrovascular disease by evaluating factors reportedly associated with upper limb function recovery.