In addition to conventional rehabilitation programs (therapy-as-usual), the subject in the no-splint group received 1-hour task-oriented training, totally 15 times in a duration of one month. The content of task-oriented training contained picking cubes from side to side with three-jaw grasp, lifting the cone to the shoulder height with the cylindrical grip and an extended elbow, picking up pegs and inserting it into the hole with palmar pinch, and grasping a soap to simulate wiping the body. Among the four tasks, two were chosen to train the subjects according to the abilities of subjects.

A participant in sitting position with the palm of the affected side placed at the edge of table was instructed to actively move his or her fingers and wrists as much as possible. Then the AROM of the index finger and wrist were measured by a goniometer.

A participant in sitting position with the palm of the affected side placed at the edge of table was instructed to actively move his or her fingers and wrists as much as possible. Then the AROM of the index finger and wrist were measured by a goniometer.

For the grip strength measurement, a participant in sitting position was instructed to squeeze a dynamometer with the unaffected hand as hard as he or she can for about 3 seconds. Then the hand of affected side was tested in the same manner. Three consecutive measurements with a 3-minute interval were performed for each hand and the arithmetic mean value of the 3 trials was used for statistical analysis.

For the grip strength measurement, a participant in sitting position was instructed to squeeze a dynamometer with the unaffected hand as hard as he or she can for about 3 seconds. Then the hand of affected side was tested in the same manner. Three consecutive measurements with a 3-minute interval were performed for each hand and the arithmetic mean value of the 3 trials was used for statistical analysis.

For the grip strength measurement, a participant in sitting position was instructed to squeeze a dynamometer with the unaffected hand as hard as he or she can for about 3 seconds. Then the hand of affected side was tested in the same manner. Three consecutive measurements with a 3-minute interval were performed for each hand and the arithmetic mean value of the 3 trials was used for statistical analysis.

Using the standardized questions from the upper-extremity MAL, a participant was rated how much (Amount of Use scale, MAL-AOU) and how well (Quality of Movement scale, MAL-QOM) he or she used the upper limb of affected side to accomplish each functional activity during the past week. Both scales were anchored at 6 points (AOU scale: 0 = not used, 5 = the same as before stroke or brain injury; QOM scale: 0 = not used, 5 = normal).17, 18 A mean MAL score is the mean of all item scores.

Using the standardized questions from the upper-extremity MAL, a participant was rated how much (Amount of Use scale, MAL-AOU) and how well (Quality of Movement scale, MAL-QOM) he or she used the upper limb of affected side to accomplish each functional activity during the past week. Both scales were anchored at 6 points (AOU scale: 0 = not used, 5 = the same as before stroke or brain injury; QOM scale: 0 = not used, 5 = normal).17, 18 A mean MAL score is the mean of all item scores.

Using the standardized questions from the upper-extremity MAL, a participant was rated how much (Amount of Use scale, MAL-AOU) and how well (Quality of Movement scale, MAL-QOM) he or she used the upper limb of affected side to accomplish each functional activity during the past week. Both scales were anchored at 6 points (AOU scale: 0 = not used, 5 = the same as before stroke or brain injury; QOM scale: 0 = not used, 5 = normal).17, 18 A mean MAL score is the mean of all item scores.

For the Fugl-Meyer Upper Extremity Assessment (FMA), a 3-graded scale with 0 as minimum and 2 as maximum was applied for grading of sensorimotor function.19 The testing procedure was conducted in a standardized manner according to the written instructions originally published and some additional general guidelines.20, 21 For motor function assessment, the Fugl-Meyer Assessment for upper extremity (FMA-UE) was used. Clear and precise instructions were given to a seated participant for each movement. The unaffected side simulated first and then the affected side was evaluated. Each activity was repeated 3 times and the highest score was recorded. The maximum total score was 66. For sensory function assessment, the Modified Fugl-Meyer sensory assessment (FMA-sensory) was used. This included tests of the light touch, temperature, tactile localization and position sensation of the upper arm, forearm, hand, thigh, calf and foot of the affected side. The maximum total score was 44.

For the Fugl-Meyer Upper Extremity Assessment (FMA), a 3-graded scale with 0 as minimum and 2 as maximum was applied for grading of sensorimotor function.19 The testing procedure was conducted in a standardized manner according to the written instructions originally published and some additional general guidelines.20, 21 For motor function assessment, the Fugl-Meyer Assessment for upper extremity (FMA-UE) was used. Clear and precise instructions were given to a seated participant for each movement. The unaffected side simulated first and then the affected side was evaluated. Each activity was repeated 3 times and the highest score was recorded. The maximum total score was 66. For sensory function assessment, the Modified Fugl-Meyer sensory assessment (FMA-sensory) was used. This included tests of the light touch, temperature, tactile localization and position sensation of the upper arm, forearm, hand, thigh, calf and foot of the affected side. The maximum total score was 44.

For the Fugl-Meyer Upper Extremity Assessment (FMA), a 3-graded scale with 0 as minimum and 2 as maximum was applied for grading of sensorimotor function.19 The testing procedure was conducted in a standardized manner according to the written instructions originally published and some additional general guidelines.20, 21 For motor function assessment, the Fugl-Meyer Assessment for upper extremity (FMA-UE) was used. Clear and precise instructions were given to a seated participant for each movement. The unaffected side simulated first and then the affected side was evaluated. Each activity was repeated 3 times and the highest score was recorded. The maximum total score was 66. For sensory function assessment, the Modified Fugl-Meyer sensory assessment (FMA-sensory) was used. This included tests of the light touch, temperature, tactile localization and position sensation of the upper arm, forearm, hand, thigh, calf and foot of the affected side. The maximum total score was 44.

An electrophysiological examination was performed with an active recording surface electrode (G1) placed over the belly of the abductor digiti minimi of the affected side, a reference electrode (G2) placed on the phalanx of the little finger, and a ground electrode placed on the back of the hand between the stimulating and recording electrodes. Supra-maximal stimulations at the wrist 8 cm away from G1 was performed to obtain the compound muscle action potential (CMAP) and F wave of the ADM muscle with the cathode distal and proximal to the anode, respectively. The stimulation was given less than 1 Hz to avoid influences from the previous stimulus. The ratio of amplitude of the mean and the maximal F-wave to the CMAP (F/M ratio) were then calculated.

An electrophysiological examination was performed with an active recording surface electrode (G1) placed over the belly of the abductor digiti minimi of the affected side, a reference electrode (G2) placed on the phalanx of the little finger, and a ground electrode placed on the back of the hand between the stimulating and recording electrodes. Supra-maximal stimulations at the wrist 8 cm away from G1 was performed to obtain the compound muscle action potential (CMAP) and F wave of the ADM muscle with the cathode distal and proximal to the anode, respectively. The stimulation was given less than 1 Hz to avoid influences from the previous stimulus. The ratio of amplitude of the mean and the maximal F-wave to the CMAP (F/M ratio) were then calculated.

An electrophysiological examination was performed with an active recording surface electrode (G1) placed over the belly of the abductor digiti minimi of the affected side, a reference electrode (G2) placed on the phalanx of the little finger, and a ground electrode placed on the back of the hand between the stimulating and recording electrodes. Supra-maximal stimulations at the wrist 8 cm away from G1 was performed to obtain the compound muscle action potential (CMAP) and F wave of the ADM muscle with the cathode distal and proximal to the anode, respectively. The stimulation was given less than 1 Hz to avoid influences from the previous stimulus. The ratio of amplitude of the mean and the maximal F-wave to the CMAP (F/M ratio) were then calculated.