Patients and methods: This prospective, randomized-controlled study included 24 unilateral hemiplegic patients (10 males, 14 females; mean age 64.1±14.8 years; range 22 to 84 years) with GHS as assessed by ultrasonography between December 2013 and September 2014. The patients were randomly divided into two groups as those in the NMES group (n=12) who were treated with NMES to supraspinatus, upper trapezius, and posterior deltoid muscles combined with conventional physiotherapy and as those in the control group (n=12) who were received conventional physiotherapy alone. Clinical (the Brunnstrom Motor Recovery Stage, Visual Analog Scale [VAS] for Pain and Shoulder Disability Questionnaire [SDQ]) and ultrasonographic (acromion-greater tuberosity distance, thicknesses of supraspinatus, upper trapezius, and posterior deltoid muscles) variables were evaluated before and after treatment in both groups.

Results: The SDQ index, acromion-greater tuberosity distance, and supraspinatus muscle thickness were improved in the NMES group, compared to the control group (for all p<0.05). The VAS-pain scores decreased in both groups. There was no statistically significant alterations in the other measurements in both groups (for all p>0.05). The percentage change (%) of the VAS-pain scores was not significantly different between two groups (p=0.03).

Conclusion: Our study results showed that GHS decreased after 20 sessions of NMES treatment. Based on these findings, ultrasonography appears to be a proper imaging tool for the evaluation of GHS in stroke patients. Keywords : Electrical stimulation; glenohumeral subluxation; ultrasonography