Background and Purpose : The hemiplegic upper extremity is affected in many stroke patients, and recovery is often poor. The purpose of this study was to assess the efficacy of electroacupuncture (EA) in enhancing the upper extremity motor and functional recovery of ischemic stroke patients. Subjects and Methods : Forty ischemic stroke patients (the upper extremity Fugl-Meyer motor scale (FM) score lower than 46, lesion location within middle cerebral artery territory) within 2 weeks of stroke onset were randomly allocated to either an EA group that received EA treatment or a control group that received only routine ward care. The EA was applied at Quchi-Shousanli (LI11-LI10), Waiguan-Hegu (TE5-LI4) points on the hemiparetic side six times per week for 4 weeks. The frequency of stimulation was 25-50Hz and the intensity was set at a level sufficient to induce muscle contraction. EA treatment was given for 30 minutes and all patients of both groups received standard rehabilitation program. Outcomes were assessed, in a blind manner, before treatment began and at 4 weeks after treatment, with the FM, the Motor Power score (MP) for shoulder/elbow, and the subsection of the Modified Barthel Index (MBI) for drinking/feeding/dressing upper body/grooming. Results : These 2 groups had comparable clinical characteristics, lesion location, lesion size, and pretreatment impairment scores. By the end of treatment, the EA group showed significantly more improvement than the control group in the subsection of the FM for shoulder/elbow/coordination (6.4 vs. 3.7; P=0.047) and the MP for shoulder/elbow (5.3 vs. 3.3; P=0.008). The subsection scores of the MBI for drinking/feeding/dressing upper body/grooming were not significantly different between two groups. No adverse effects due to treatment were found Conclusion : These results suggest that EA enhances the upper extremity motor recovery of acute stroke patients. However, this study failed to demonstrate any significant functional benefit related with upper extremity. Future study should be carried out in a larger sample size and use the functional outcome measure that is more specific and sensitive to the upper extremity.