Prediction of Gait without Physical Assistance after Inpatient Rehabilitation in Severe Subacute Stroke Subjects

Gait recovery is one of the main therapeutic goals of the rehabilitation for patients after a stroke.

This study is aimed at describing the frequency of achievement of gait without physical assistance in poststroke subacute patients by the time of discharge from a rehabilitation hospitalized program. Secondarily, our goal is to identify gait without physical assistance predictors in this same population based on the admission's clinical and demographic conditions.

Data from 185 first unilateral hemispheric stroke patients that need physical assistance to walk at admission were analyzed. The sample was dichotomized into gait with physical assistance and gait without physical assistance to calculate the frequency of achievement of gait without assistance at discharge. Multivariate logistic modeling was applied to identify prognostic factors for regaining gait without physical assistance.

Gait without assistance was achieved in 50.27% of the subjects. Five variables were identified for the prediction model: age (Odds ratio [OR] = .87, 95% confidence interval [CI] = .83-.92), gender (OR = .37, 95% CI = .14-.94), time between stroke and hospitalization (OR = .96, 95% CI = .94-.99), initial Berg Balance (OR = 1.52, 95% CI = 1.23-1.88), and initial lower limb Fugl Meyer (OR = 1.17, 95% CI = 1.07-1.27).

Although discharge planning is complex, achievement of gait without physical assistance is undoubtedly a landmark to decide on hospitalization discharge. Half of this sample was able to walk without physical assistance at hospitalization discharge. Five clinical and demographic conditions at admission were found predictors of gait without physical assistance at inpatient discharge.