In standing, people with chronic stroke typically demonstrate postural control problems, including increased sway of their center of gravity (COG), reduced ability in voluntary weight-shift, slow postural response time to perturbation impaired sensory integration, and excessive reliance on visual input to maintain an upright stance. Exercise programs targeted to improve their postural control and mobility problems usually last 4 to 10 weeks, and have incorporated muscle strengthening, transfer of positions, weight shifting, agility training on an obstacle course, and walking on a treadmill. Given their multidimensional nature, it has not been possible to delineate which exercise component is effective for improving balance control. Moreover, exercise progression and supervision by health care professionals is required, so self-practice by stroke survivors has not been feasible. The possibility of carry-over effects from such therapeutic programs has been also rarely addressed in the literature. 

Their previous studies on Tai Chi demonstrated that experienced Tai Chi practitioners had better knee joint proprioception and standing balance than control subjects similar in age and activity level. Tai Chi practitioners have shown improved control of voluntary weight-shifting, better balance in perturbed stance under visual- or vestibular-challenged conditions, as well as better balance in perturbed single-leg stance. Other investigators have shown that long-term practice of Tai Chi can reduce the risk of falls. Even 4 weeks of intensive daily Tai Chi practice can be sufficient to improve the standing balance in healthy elderly subjects, as opposed to the 4 to 10 weeks of exercise programs for stroke survivors mentioned above. This study was, therefore, designed to examine the effects of a 12-week community-based training of a short-form of Tai Chi on standing balance control after stroke.

One hundred thirty-six subjects >6 months after stroke were randomly assigned to a control group (n = 62) practicing general exercises or a Tai Chi group (n = 74) for 12 weeks of training. Each week, 1 hour of group practice was supplemented by 3 hours of self-practice. We used a short-form of Tai Chi consisting of 12 forms that require whole-body movements to be performed in a continuous sequence and demands concentration. A blinded assessor examined subjects at baseline, 6 weeks (mid-program), 12 weeks (end-program), and 18 weeks (follow-up). The 3 outcome measures were (1) dynamic standing balance evaluated by the center of gravity (COG) excursion during self-initiated body leaning in 4 directions, (2) standing equilibrium evaluated in sensory challenged conditions, and (3) functional mobility assessed by Timed-up-and-go score. Mixed model repeated-measures analysis of variance was used to examine between-group differences. 

When compared with the controls, the Tai Chi group showed greater COG excursion amplitude in leaning forward, backward, and toward the affected and nonaffected sides (P < .05), as well as faster reaction time in moving the COG toward the nonaffected side (P = .014) in the end-program and follow-up assessments. The Tai Chi group also demonstrated better reliance on vestibular integration for balance control at end-program (P = .038). However, neither group improved significantly in Timed-up-and-go scores. 

Twelve weeks of short-form Tai Chi produced specific standing balance improvements in people with chronic stroke that outlasted training for 6 weeks.