A cohort study on longitudinal changes in postural balance during the first year after stroke

Abstract

Introduction: Many patients with strokes report increased incidence of fall that can be due to impaired postural balance. The recovery of balance in patients with varying degrees of impairments and activity limitations is less studied, and whether individuals with mild paresis can recover their balance faster is unclear. Better knowledge about factors influencing the recovery of postural balance can be used to guide clinical management after stroke to provide the right rehabilitation to the right person at the right time, and thus to avoid potential fall incidences.

Objective: This study aims to examine longitudinal changes in postural balance during the first year after stroke.

Methods: Postural balance was assessed using the Berg Balance Scale (BBS) within 5 days, 1, 2, and 3 months and 1-year post-stroke. Stroke severity was stratified using a cluster analysis by including multidimensional baseline measures. A longitudinal mixed-effect model was constructed to analyze changes in proportional balance impairment by stroke severity over time. Individuals with a cut-off of BBS below 45 scores were identified through a classification algorithm using baseline predictors.

Results: A total of 135 patients were stratified to mild stroke (77 [57%] patients) or moderate stroke (58 [43%] patients). Ninety-three patients were included in the longitudinal analysis. Significant recovery was found at 1-year for moderate stroke (48% recovery from the initial impaired postural balance, adjusted P < 0.001), but not for mild stroke, after adjusting for age and cognition. Both stroke severities had a maximal recovery in postural balance at 3 months post-stroke, but the moderate stroke group deteriorated after that. Patients with higher age and worse cognition had more severe balance impairments. The classification model achieved a sensitivity of 0.95 (95% confidence interval [CI]: 0.91-0.98) and a specificity of 0.99 (95% CI: 0.98-1.0) for classifying individuals with BBS below 45 points.

Conclusions: This study indicates that continuous improvements in postural balance ends at 3 months regardless for mild or moderate stroke groups, and patients with moderate stroke significantly deteriorate in postural balance after 3 months.

Keywords: Berg Balance Scale; Cerebrovascular Accident; Impairment of Postural Balance; Longitudinal Analysis; Stroke Recovery.

Fig. 1

Variable importance of baseline variables in the cluster analysis. The corresponding domain of each variable is indicated following the framework of The International Classification of Functioning, Disability, and Health. The variable importance was derived from the mean square error with higher values indicating higher importance. Two clusters were determined as an optimal number of clusters on the basis of silhouette width (0.48). Stability of the clusters was assessed using the Jaccard similarity through resampling of the data 500 times. The Jaccard similarity was 0.97, which indicates stable clusters. BBS, Berg Balance Scale; BI, Barthel Index; FMA, Fugl-Meyer Assessment; HADS, Hospital Anxiety and Depression Scale; NIHSS, National Institutes of Health Stroke Scale; IQR, interquartile range; LE, lower extremity; MoCA, Montreal Cognitive Assessment; mRS, modified Rankin Scale; ROM, passive joint motion; SD, standard deviation; TUG, timed up-and-go test; UE, upper extremity

Fig. 2

A Estimated means of proportional balance impairments and 95% confidence intervals across five time points by stroke severity. B Individual changes in proportional balance impairments by stroke severity, cognition, and age between 3 months and 1-year post-stroke. Difference ≥ 0 indicates an increase in balance impairment, whereas < 0 is a decrease in impairment. A total of 35 patients ≥ 75 years of age had an increase in balance impairment from 3 months to 1-year post-stroke. ^a Estimated means were converted from least square means after adjusting for age and cognition. MoCA, Montreal Cognitive Assessment

Fig. 3

Longitudinal Berg Balance scale (BBS) in each individual by risk of falling in mild and moderate stroke. A cut-off of < 45 points, across any time point during the first year indicates individuals with a high risk of falling

Fig. 4

Mean decrease in accuracy after permutation of each variable in the random forest model. FMA-LE, BI-transfers, and age are the three most contributing variables for the model performance.BI, Barthel Index; FMA, Fugl-Meyer Assessment; HADS, Hospital Anxiety and Depression Scale; NIHSS, National Institutes of Health Stroke Scale; IQR, interquartile range; LE, lower extremity; MoCA, Montreal Cognitive Assessment; mRS, modified Rankin Scale; ROM, passive joint motion; SD, standard deviation; UE, upper extremity