Association of postural balance and falls in adult patients receiving haemodialysis: A prospective cohort study

Abstract Background Static postural balance performance is often impaired in people receiving haemodialysis (HD) for the treatment of stage-5 chronic kidney disease (CKD-5). However, the question as to whether lower postural balance is associated with adverse clinical outcomes such as falls has not been addressed yet. Research question We conducted a prospective cohort study to explore the association between static postural balance and falls in people receiving HD. We hypothesised that higher postural sway would be associated with increased odds of falling. Methods Seventy-five prevalent CKD-5 patients receiving HD (age: 61.8 ± 13.4 years) from three Renal Units were enrolled in this prospective cohort study. At baseline, postural balance was assessed with a force platform in eyes open (EO) and eyes closed (EC) conditions. Centre of pressure (CoP) measures of range, velocity and area were taken for the analysis. Falls experienced by study participants were prospectively recorded during 12 months of follow-up. Secondary outcomes included timed-up and go, five-repetition sit-to-stand test and the Tinetti falls efficacy scale (FES). Results In multivariable logistic regression analysis, higher CoP range in medial-lateral direction during EC was associated with increased odds of falling (OR: 1.04, 95 %CI: 1.00−1.07, p = 0.036). In ROC curve analysis, CoP velocity in EO exhibited the greatest prognostic accuracy (AUC: 0.69, 95 %CI: 0.55−0.82), however this was not statistically different from CoP measures of area and range. None of the postural balance measures exceeded the prognostic accuracy of the FES (AUC: 0.70, 95 %CI: 0.58−0.83, p = 0.005). Significance This prospective cohort study showed that higher postural sway in medial-lateral direction was associated with increased odds of falling in people receiving HD. CoP measures of range, velocity and area displayed similar prognostic value in discriminating fallers from non-fallers. The overall utility of static posturography to detect future fall-risk may be limited in a clinical setting.